REVIEW

Arthropod segmentationErik Clark1,2,*, Andrew D. Peel3 and Michael Akam2

ABSTRACTThere is now compelling evidence that many arthropods pattern theirsegments using a clock-and-wavefront mechanism, analogous to thatoperating during vertebrate somitogenesis. In this Review, wediscuss how the arthropod segmentation clock generates arepeating sequence of pair-rule gene expression, and how this isconverted into a segment-polarity pattern by ‘timing factor’wavefrontsassociated with axial extension. We argue that the gene regulatorynetwork that patterns segments may be relatively conserved,although the timing of segmentation varies widely, and double-segment periodicity appears to have evolved at least twice. Finally,we describe how the repeated evolution of a simultaneous(Drosophila-like) mode of segmentation within holometabolaninsects can be explained by heterochronic shifts in timing factorexpression plus extensive pre-patterning of the pair-rule genes.

KEY WORDS: Arthropods, Segmentation, Patterning, Pair-rulegenes, Drosophila, Tribolium

IntroductionArthropods are an ecdysozoan phylum defined by their segmentedbodies and jointed limbs. True arthropods (euarthropods) comprisethree living clades: Chelicerata (spiders, scorpions and mites),Myriapoda (centipedes and millipedes), and Pancrustacea(crustaceans and insects). The closest relatives of arthropods areonychophorans (velvet worms) and tardigrades (water bears); togetherthese phyla form the segmented superphylum Panarthropoda(Fig. 1A).The great diversity of arthropod species is testament to the

evolutionary potential of a segmented body plan: a modularorganisation of fundamentally similar units arrayed serially alongthe anteroposterior (AP) axis (Hannibal and Patel, 2013). Arthropodsegments, and their associated appendages, have diversifiedremarkably through adaptation to specific functions, such asfeeding, locomotion or reproduction. In addition, segment numbercan vary enormously, from fewer than 20 in insects andmalacostracan crustaceans, to over 100 in certain centipedes andmillipedes, resulting in a wide spectrum of organismal forms(Brusca et al., 2016). With over a million named species, arthropodshave colonised and exploited almost every environment on Earth,thanks in no small part to the evolution of segmentation.

Our understanding of how segments are patterned in arthropodembryos has traditionally been heavily influenced by study of thefruit fly Drosophila melanogaster. Over the past two decades,research into sequentially segmenting species has complementedthe well-establishedDrosophilamodel, resulting in the discovery ofan arthropod ‘segmentation clock’, and an outline of conserved anddivergent aspects of arthropod segment patterning networks. In thelight of these findings, recent studies have re-examinedsegmentation in Drosophila, uncovering new subtleties andinterpreting their evolutionary significance.

In the sections that follow, we provide a general overview ofarthropod segmentation and review our current understanding ofthree key issues: (1) the nature of the arthropod segmentation clock;(2) how the ‘pair-rule’ genes pattern segments; and (3) the evolutionof Drosophila-style simultaneous segmentation from a sequentiallysegmenting ancestral state. We also reflect on the origins ofarthropod segmentation (Box 1) and the control of segment number(Box 2). As we have chosen to focus on the time window whensegments are actively being patterned, we do not discuss earlier APpatterning processes, such as axis specification, or later ones, suchas segment morphogenesis.

Overview of arthropod segmentationSegments and parasegmentsIn arthropods, morphological segmentation is built upon a morefundamental developmental unit, the ‘parasegment’ (Martinez-Arias and Lawrence, 1985). Parasegment boundaries are establishedduring embryogenesis by ‘segment-polarity’ genes, such asengrailed and wingless, which are expressed in a series ofpersistent stripes along the AP axis. Interestingly, parasegmentsare offset slightly from morphological segments: parasegmentboundaries fall at the anterior edge of each engrailed domain andline up with the middle of each appendage, whereas segmentboundaries fall at the posterior edge of each engrailed domain andlie in between the appendages (Fig. 1B). Analogous to vertebrate‘resegmentation’ (each vertebra being formed from portions of twodifferent somite pairs), this developmental phase shift makes senseif the role of the parasegments is chiefly to organise the nervoussystem and associated appendicular structures, whereas the role ofmorphological segmentation is to protect these centres and formexoskeletal articulations between them (Deutsch, 2004).

Each segment-polarity gene is expressed at a particular positionwithin a segmental unit, and the overall arrangement is remarkablyconserved across Panarthropoda (Damen, 2002; Janssen and Budd,2013). A central goal of segmentation research is to understand howupstream regulatory processes establish this important patternwithin the embryo.

Sequential segmentation and the segment addition zoneMost arthropods pattern their segments sequentially, from head totail, coupling the segmentation process to progressive axialextension (Sander, 1976). They usually specify some number ofanterior segments in the blastoderm, but the majority of the

1Department of Systems Biology, Harvard Medical School, Boston, MA 02115,USA. 2Department of Zoology, University of Cambridge, Cambridge, CB2 3EJ, UK.3School of Biology, Faculty of Biological Sciences, University of Leeds, Leeds, LS29JT, UK.

*Author for correspondence (erik_clark@hms.harvard.edu)

E.C., 0000-0002-5588-796X; A.D.P., 0000-0002-9914-3508; M.A., 0000-0003-0063-2297

This is an Open Access article distributed under the terms of the Creative Commons AttributionLicense (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use,distribution and reproduction in any medium provided that the original work is properly attributed.

1

© 2019. Published by The Company of Biologists Ltd | Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

segments emerge rhythmically from a posterior ‘segment additionzone’ (SAZ) after the blastoderm-to-germband transition. The SAZretracts posteriorly as new segments are added to the trunk,generally shrinking in size, until the embryo reaches full germbandextension (Fig. 1C).‘SAZ’ is now preferred over the traditional term ‘growth zone’,

because it makes no assumption of localised and continuous cellproliferation in the posterior of the embryo (Janssen et al., 2010).The material for new segments is generally provided by acombination of cell division and convergent extension, but – as invertebrates – the relative contributions of these cell behaviours toaxial elongation vary widely across species (Auman et al., 2017;Benton, 2018; Benton et al., 2016; Mito et al., 2011; Nakamotoet al., 2015; Steventon et al., 2016). Accordingly, although celldivision may in some species be coordinated with segment addition,segment patterning processes do not appear to be mechanisticallydependent on the cell cycle (Cepeda et al., 2017), aside from inspecial cases such as malacostracan crustaceans. This group exhibitsa highly derived mode of segmentation in which patterning occursthrough regimented asymmetrical divisions of rows of posteriorcells (Scholtz, 1992).Although the shape, size and proportions of the SAZ vary

considerably across species, certain features are conserved.Segment-polarity stripes emerge at the anterior of the SAZ, and

Wnt is expressed at its posterior (Williams and Nagy, 2017).Between these limits, we define the ‘anterior SAZ’ as the portion ofthe SAZ that contains segments in the process of being patterned,and the ‘posterior SAZ’ as the portion that contains cells not yetassigned to any particular prospective segment. These functionallydefined regions correlate with the differential expression of keydevelopmental transcription factors; for example, Caudal (thearthropod homologue of the vertebrate Cdx proteins) appears tobe specifically associated with the posterior SAZ (Auman et al.,2017; Clark and Peel, 2018).

Importantly, SAZ identity is transient and dynamic for any givencell. With the generation of each new segment, newly patternedtissue ‘leaves’ the anterior SAZ, which is simultaneously‘replenished’ by cells from the posterior SAZ. (Whether cellsflow anteriorly out of the SAZ or the SAZ retracts posteriorly alongthe embryo depends on one’s choice of reference frame.) Thus, acell that starts out within the posterior SAZ, expressing one set ofgenes, will at some point end upwithin the anterior SAZ, expressinga different set of genes, and finally within the segmented germband,expressing yet another (Fig. 1C). This provides a mechanisticexplanation for the tight coupling between axial elongation andthe segmentation process, because the changing expressionlevels of SAZ-associated factors such as Caudal are likely totrigger coordinated expression changes in segment patterning

Drosophila melanogaster

Anopheles gambiae

Bombyx mori

Diptera

Coleoptera

Lepidoptera

Hymenoptera

Holometabola

Hemiptera

Blattodea

Orthoptera

Insecta

Oncopeltus fasciatus

Tribolium castaneum

Nasonia vitripennis

Apis mellifera

Artemia franciscana

Periplaneta americana

Schistocerca sp.

Gryllus bimaculatus

Strigamia maritima

Parhyale hawaiensis

Daphnia magna

Parasteatoda tepidariorum

Glomeris marginata

Lithobius forficatus

Hypsibius dujardini

Euperipatoides rowelli

Cupiennius salei

Pancrustacea

Euarthropoda

Tardigrada

Onychophora

Malacostraca

Chelicerata

Myriapoda

Branchiopoda

Panarthropoda

Fruit fly

Mosquito

Silkmoth

Milkweed bug

Flour beetle

Jewel wasp

Honeybee

Brine shrimp

Cockroach

Grasshopper

Cricket

(Geophilomorph centipede)

(Amphipod crustacean)

Water flea

House spider

Pill millipede

Brown centipede

Tardigrade

Velvet worm

Wandering spider

Dermestes maculatus Hide beetle

Embryo

Adult

Parasegment

Segment

Parasegmentboundary

A B

C

Anterior SAZ

PosteriorSAZ

A P

A P

A P

Segmentedgermband

Axial extension

Pair-rule patterning UnclearKey

Fig. 1. Overview of arthropod segmentation. (A) Phylogenetic tree of notable arthropod model species (based on Misof et al., 2014; Schwentner et al., 2017).Red text indicates species known to use pair-rule patterning; the status of Oncopeltus is currently unclear. Branch lengths not to scale. (B) Diagram showing therelationship between parasegments and segments. Pink represents engrailed expression. A, anterior; P, posterior. (C) Schematic time series of an arthropodembryo undergoing sequential segmentation. engrailed stripes (pink) emerge sequentially from a retracting segment addition zone (SAZ, blue) as the germbandextends posteriorly. Green dots mark the progress of a specific individual cell that starts in the posterior SAZ (dark blue), transiently forms part of the anterior SAZ(light blue), and ends up in the segmented germband.

2

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

genes as the SAZ retracts (Clark and Peel, 2018; El-Sherifet al., 2014).

Segment patterning by a clock-and-wavefront mechanismArthropod segmentation is frequently compared to vertebratesomitogenesis (reviewed by Hubaud and Pourquié, 2014; Oateset al., 2012). Although segments and somites are not homologousmorphological structures, it is now becoming clear that botharthropods and vertebrates have converged on a ‘clock-and-wavefront’ strategy (Cooke and Zeeman, 1976) to pattern their APaxis. Temporal periodicity is generated by an oscillator (the ‘clock’),and progressively translated into spatial periodicity by a secondsignal (the ‘wavefront’), which travels along an axis and freezes(or reads out) the phase of the clock.In vertebrates, the clock consists of cycles of gene expression in

the presomitic mesoderm (PSM), whereas in arthropods it consistsof cycles of gene expression in the posterior ectoderm. In both thevertebrate anterior PSM and the arthropod anterior SAZ, theoscillations are slowed by the retraction of posterior signalsassociated with axial extension, converting them into a series ofstripes. These stripes then pattern other genes, which determine theAP polarity of somites (in vertebrates) or segments (in arthropods).Curiously, the periodicity of the segmentation clock is not fixed

across arthropods. Most groups pattern a single new segment foreach cycle of the clock (as do vertebrates), but some species patterntwo segments in each cycle, meaning that their clocks have adouble-segment (or ‘pair-rule’) periodicity (Chipman et al., 2004;Sarrazin et al., 2012).

Other modes of segmentationThe sequential mode of segmentation is widespread and almostcertainly ancestral within arthropods. However, across species thetiming of segmentation can vary dramatically relative to otherdevelopmental events.For example, arthropod embryos differ widely in the number of

segments they pattern at the blastoderm stage, versus afterwardsduring germband extension. In insects, this variation is roughlycorrelated with a spectrum of ‘germ types’ defined in the pre-molecular era (Davis and Patel, 2002; Krause, 1939), but forsimplicity and generality, we have chosen to eschew suchterminology in this Review. Instead, we will refer to sequentialsegmentation (usually occurring in a germband, under the control ofa segmentation clock) versus simultaneous segmentation (usuallyoccurring in a blastoderm, downstream of non-periodic spatialcues). The mechanisms underlying simultaneous segmentation arediscussed in more detail below.Outside of the insects, many arthropod groups undergo post-

embryonic segmentation, i.e. delay the development of a portion ofthe AP axis until after hatching. In crustaceans with naupliar larvae,for example, only the head segments are patterned in the embryo,and trunk segments develop sequentially from a SAZ-like regionafter the larva has begun feeding (Anderson, 1973). Other, lessextreme, examples are found within myriapods: these pattern thehead and the first trunk segments in the embryo, but may add one ormore trunk segments after each moult (Blower, 1985).Our focus here is on the segmentation of the trunk (i.e. the axial

patterning of the gnathal, thoracic and abdominal segments), butnote that there are other parts of the arthropod body that aresegmented by different mechanisms, such as the anterior head(Posnien et al., 2010) or the jointed appendages (Angelini andKaufman, 2005a). Within the trunk itself, the mechanisms wedescribe specifically control ectodermal segmentation; mesodermal

segmentation occurs later, apparently directed by inductive signalsfrom the segmented ectoderm (Azpiazu et al., 1996; Green andAkam, 2013; Hannibal et al., 2012). Finally, there is evidence thatdorsal segmentation in millipedes is decoupled from ventralsegmentation, which later leads to segment fusions (Janssen,2011; Janssen et al., 2004).

Segment patterning genesMost of the arthropod segmentation genes we know about wereoriginally identified from a genetic screen in Drosophila (Nüsslein-Volhard and Wieschaus, 1980). Drosophila represents an extremeexample of simultaneous segmentation, patterning all but its mostterminal segments in the blastoderm. It has taught us a lot about howsegmentation genes regulate one another’s expression (Akam,1987; Nasiadka et al., 2002), but studies in other arthropods were(and are) necessary to reveal how these networks relate to moreancestral modes of segmentation (Peel et al., 2005).

InDrosophila, as in other arthropods, the segment-polarity genesare patterned by the pair-rule genes, which code for varioustranscription factors. In Drosophila, the pair-rule genes areexpressed in stripes in the blastoderm, but in sequentiallysegmenting species they are also expressed in the SAZ (Patelet al., 1994). In general, the pair-rule genes that turn on earliest inDrosophila (‘primary’ pair-rule genes) are expressed in theposterior SAZ in sequentially segmenting species, and may

Box 1. The evolutionary origins of arthropodsegmentationThe major segmented phyla – arthropods, annelids and chordates – areevolutionarily distant and separated by many unsegmented groups.Although losses of segmentation are possible in evolution (e.g. fromspoon worms and peanut worms within annelids), we are sceptical aboutthe existence of a segmented urbilaterian ancestor that could have givenrise to all three phyla (Couso, 2009). Instead, segmentation appears tohave evolved repeatedly during animal evolution, involving variousdevelopmental mechanisms (Graham et al., 2014).Someof the developmental commonalities between different segmented

phyla may reflect bilaterian homologies that predate segmentation itself,such as elongation of the body from a posterior zone (Jacobs et al., 2005;Martin and Kimelman, 2009). Other similarities may reflect the convergentadoption of generic patterning strategies, such as molecular oscillators(Richmond and Oates, 2012). Finally, certain similarities may reflect theparallel redeployment of ancient molecular mechanisms (Chipman, 2010),and therefore require both homology and convergence to fully explain. Forexample, segment boundary formation in some, but not all, annelids showsstriking similarities to parasegment boundary formation in arthropods (Drayet al., 2010; Prud’homme et al., 2003; Seaver et al., 2001; Seaver andKaneshige, 2006). Probably, this boundary specification mechanismevolved before trunk segmentation, possibly in the context of patterningthe head and anterior nervous system (Vellutini and Hejnol, 2016).The evolutionary success of segmented phyla emphasises the

adaptive value of diversified metameric structures, but it does notexplain why segmentation evolved in the first place. One long-standinghypothesis stresses the advantages of a segmented body for generatingcoordinated waves of muscular activity to drive locomotion (Clark, 1964).Given that most of the earliest arising segmented lineages have manysimilar segments, this seems a likely explanation for the initial origins ofserial repetition along the body axis, which was likely the forerunner formetameric segmentation. Under this scenario, repetition would beexpected first in the nervous system and body wall musculature.Interestingly, onychophorans have distinct mesodermal somites, andshow clear parasegmental boundaries in the limbs and nervous system(Eriksson et al., 2009), but show no obvious segmentation of the bodywall ectoderm.

3

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

oscillate, whereas those that turn on later (‘secondary’ pair-rulegenes) are expressed in the anterior SAZ. The periodicity of pair-rule gene expression can be segmental or double-segmentaldepending on the species (in Drosophila it is double-segmental,hence the term ‘pair-rule’), but the genes are always referred to asthe ‘pair-rule genes’ regardless. There has been some confusionover the years as to which Drosophila pair-rule genes should beclassed as primary and which as secondary or even tertiary.However, the most recent analysis (Schroeder et al., 2011), whichclassifies only paired ( prd) and sloppy paired (slp) as secondary,and all of hairy, even skipped (eve), runt, odd skipped (odd) andfushi tarazu ( ftz) as primary, meshes well with the comparativeevidence.In Drosophila, the primary pair-rule genes are patterned by the

‘gap’ genes, which code for another set of transcription factors. InDrosophila, these genes are expressed in broad, partiallyoverlapping domains along the length of the blastoderm, but insequentially segmenting species some portion of this pattern isgenerated over time, in the SAZ (Box 2). Gap genes in sequentiallysegmenting species do not seem to be important for directing pair-rule gene expression. They do, however, appear to play a relativelyconserved role in patterning the Hox genes, which regulate segmentidentity (Hughes and Kaufman, 2002a; Marques-Souza et al., 2008;Martin et al., 2016).

Nature of the arthropod segmentation clockOscillating gene expression in the SAZSome segmentation genes exhibit extremely variable expressionpatterns in the posterior SAZs of fixed embryos, suggesting thatthey continually turn on and off over time. In the beetle Tribolium,split-embryo experiments have confirmed that this variabilityresults from a temporally dynamic ‘segmentation clock’ withinindividuals rather than spatially variable expression betweenindividuals (Sarrazin et al., 2012). Expression dynamicity has alsobeen demonstrated in Tribolium by comparing the average patternsof finely staged cohorts of embryos, by visualising discrepanciesbetween the transcript and protein domains of a given gene, and bygaining an understanding of cell dynamics within the SAZ via liveimaging (Benton, 2018; El-Sherif et al., 2012; Sarrazin et al., 2012).In other species, gene expression dynamics within the SAZ haverarely been studied in detail. However, convincing ‘pseudo time-series’ assembled from carefully staged Strigamia (centipede) andParasteatoda (spider) embryos imply that oscillatory dynamics arewidespread (Brena and Akam, 2013; Schönauer et al., 2016).Candidate gene approaches in species including Tribolium,

Strigamia, the millipede Glomeris, and a second spider,Cupiennius, indicate that oscillating SAZ genes include theprimary pair-rule genes hairy, eve, runt and odd (Choe et al.,2006; Damen et al., 2005; Green and Akam, 2013; Janssen et al.,2011). [The segmentation role of ftz is less widely conserved (Pick,2016).] In addition, Notch signalling components appear to oscillatein many clades (see below), as do prd and hedgehog in spiders(Davis et al., 2005; Schoppmeier and Damen, 2005a; Schwager,2008). However, as there has not yet been an exhaustive screen forcyclic expression, we do not know how many other genes may havebeen missed.Measurements from Tribolium (El-Sherif et al., 2012; Nakamoto

et al., 2015; Sarrazin et al., 2012) and Strigamia (Brena and Akam,2012) suggest an oscillation period in these species of ∼3 h at18-20°C (or equivalently ∼6 h at 13°C or ∼1.5 h at 30°C, assegmentation speed scales with developmental rate). Adjusted fortemperature, these numbers are comparable to the fastest

segmenting vertebrates, such as zebrafish or snakes (Gomez et al.,2008). Interestingly, the rate of segment addition is not constantthroughout development (Brena and Akam, 2013; Nakamoto et al.,2015). This implies that there is stage-specific variation in theoscillation period, the axial elongation rate, and/or the dynamics oftissue maturation in the SAZ (Schröter et al., 2012; Soroldoni et al.,2014).

At present, the mechanistic basis for the oscillations is not wellunderstood. Nonetheless, it is useful to think about contributingregulatory processes using a three-tier framework (Oates et al.,2012): (1) gene expression dynamics within cells; (2) signallinginteractions between cells; and (3) the changing regulatory contextalong the SAZ.

Gene expression dynamics within cellsIn vertebrates such as zebrafish, (auto)repressive interactions betweenHer/Hes transcription factors (homologues of theDrosophila pair-rulegene hairy) are thought to form the core of the segmentation clock,driving oscillations by time-delayed negative feedback (Lewis, 2003;Schröter et al., 2012). Analogously, it is possible that the arthropodsegmentation clock is driven by an intracellular negative-feedbackloop formed by some or all of the oscillating pair-rule genes.

Themain evidence for this is that knocking down primary pair-rulegenes can block segmentation and truncate the body axis, as has beenfound in Tribolium (Choe et al., 2006), the silkmoth Bombyx (Nakao,2015), a second beetle speciesDermestes (Xiang et al., 2017) and thehemipteran bug Oncopeltus (Auman and Chipman, 2018; Liu andKaufman, 2005). It can also cause the expression of other primarypair-rule genes to become aperiodic (Choe et al., 2006; Nakao,2015), suggesting that at least some of the oscillations are mutuallyinterdependent. This observation distinguishes these knockdownsfrom those of downstream patterning genes, which may also yieldasegmental phenotypes but do not perturb expression dynamics inthe SAZ (Choe and Brown, 2007; Farzana and Brown, 2008).

The topology for a pair-rule gene segmentation clock is not clear.An early RNA interference (RNAi) study in Tribolium found thateve, runt or odd knockdown resulted in truncation, whereas hairyknockdown resulted only in head defects (Choe et al., 2006). Thisled to the hypothesis that eve, runt and odd are linked into a three-gene ring circuit, and that even though hairy oscillates in the SAZ, itis not required for segmentation. Specifically, it was proposed thatEve activates runt, Runt activates odd, and Odd in turn represseseve, returning the sequence to the beginning (Fig. 2A). However,more recent evidence has raised issues with this proposal.

First, whether hairy is involved in the Tribolium segmentationclock or not remains unclear. A later study found that hairyknockdown resulted in a pair-rule phenotype for gnathal andthoracic segments (Aranda et al., 2008), and the iBeetle screen(Dönitz et al., 2015) additionally recovered posterior truncations.hairy also has a paralogue, deadpan, expressed with similardynamics in the SAZ (Aranda et al., 2008), and so its role might bemasked by functional redundancy. Finally, hairy knockdown wasrecently found to produce truncations in Dermestes (Xiang et al.,2017), and hairy is also known to regulate segment patterning in thecockroach Periplaneta (Pueyo et al., 2008), the parasitic waspNasonia (Rosenberg et al., 2014), and of course Drosophila,indicating that a role in segmentation is widely conserved.

Second, whether eve and odd are part of the primary oscillator isalso not certain. eve expression may be necessary for establishingand/or maintaining the SAZ (Cruz et al., 2010; Liu and Kaufman,2005; Mito et al., 2007; Xiang et al., 2017), and therefore its severetruncation phenotype may be independent of its potential role in the

4

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

segmentation clock. odd, on the other hand, has been found to causepair-rule and/or segment polarity defects rather than truncations inDermestes (Xiang et al., 2017) and Oncopeltus (Auman andChipman, 2018; Reding et al., 2019), although the interpretation ofthese phenotypes is complicated by the existence of odd paralogues,such as sob. Notably, neither eve nor odd shows dynamic expressionin the posterior SAZ of Oncopeltus (Auman and Chipman, 2018;Liu and Kaufman, 2005), indicating that periodicity is likely to begenerated by other genes in this species.

Finally, the specific regulatory interactions proposed for thecircuit seem unlikely. In holometabolous insects (and alsoStrigamia), eve, runt and odd are expressed sequentially withineach pattern repeat (Choe et al., 2006; Clark, 2017; Green andAkam,2013; Nakao, 2015; Rosenberg et al., 2014). In both Tribolium andBombyx, Eve is necessary for runt expression, and Runt is necessaryfor odd expression (Choe et al., 2006; Nakao, 2015). However, it isprobably not the case that Eve directly activates runt and Runtdirectly activates odd, as was proposed for Tribolium. Instead,genetic evidence from Bombyx and Drosophila (and wild-typeexpression dynamics from Tribolium) suggest something closer to a‘repressilator’ scenario (Elowitz and Leibler, 2000), where each genein the sequence represses the one before it (Fig. 2A).

In summary, although it is likely that cross-regulation plays aconsiderable role in shaping dynamic pair-rule gene expression, it isnot yet clear whether the oscillating genes are linked into a singlecircuit, whether this circuit is sufficient to generate oscillations, whatthe topology of this circuit is likely to be, nor indeed the extent towhich it may have diverged in different lineages (Krol et al., 2011).

Signalling interactions between cellsRegardless of whether the pair-rule gene network is capable ofproducing intracellular oscillations autonomously, the segmentationclock must also involve intercellular communication to keeposcillations synchronised across the SAZ. Notch signalling,known to synchronise oscillations during vertebrate somitogenesis(Liao and Oates, 2017), is the key candidate for this role. Indeed,Notch signalling components appear to oscillate along with the pair-rule genes in chelicerates (Schoppmeier and Damen, 2005b;Stollewerk et al., 2003), myriapods (Chipman and Akam, 2008;Kadner and Stollewerk, 2004), crustaceans (Eriksson et al., 2013),and some insects (Pueyo et al., 2008), suggesting that arthropodsegmentation involved Notch ancestrally.

Experiments in Cupiennius, Periplaneta, and the branchiopodcrustacean Daphnia have found that segment boundaries and theexpression of segmentation genes become disorganised when Notchsignalling is perturbed (Eriksson et al., 2013; Pueyo et al., 2008;Schoppmeier and Damen, 2005b; Stollewerk et al., 2003).Inhibiting Notch signalling also blocks segmentation (but notaxial elongation) in anostracan crustaceans (Williams et al., 2012).These findings indicate that Notch may play an explicit role ingenerating and/or coordinating pair-rule gene oscillations, perhapsvia regulation of hairy (Fig. 2B).

However, the pleiotropy of the Notch pathway means thatcharacterising this potential segmentation function may be difficult.During development, Notch signalling also regulates cellproliferation (Go et al., 1998), SAZ establishment (Chesebroet al., 2013; Oda et al., 2007; Schönauer et al., 2016), and fertility(Xu and Gridley, 2012). Accordingly, strong Notch perturbations insequentially segmenting arthropods often result in uninterpretableaxial truncations, or simply a failure to lay many eggs (Kux et al.,2013; Mito et al., 2011; Stahi and Chipman, 2016).

Eve

Odd Runt

Eve

Odd Runt

Expression within cell over time

Model 2: Repression based

Model 1: Activation based

Expression within cell over time

A Genetic ring oscillator

eveEverunt

Runtodd

Odd

eveEve

runtRunt

oddOdd

B Hypothesis: hairy links intercellular and intracellular oscillations

Eve

Odd

Runt

Hairy

Notch

Oscillator 2Oscillator 1 Functional evidence:

Periplaneta, DaphniaCupiennius

Bombyx

Drosophila

C ‘Timing factors’ coordinate segment patterning across the SAZ

A

opa caudalDichaete Wnt

P

Expression

Posterior SAZ:oscillations generate

periodicity

Anterior SAZ:the segment

pattern is resolved

Segmented germband:segment-polarity

stripes maintained

1 2 3

Fig. 2. Within-cell, between-cell and tissue-level aspects of thearthropod segmentation clock. (A) Pair-rule gene oscillations may bedriven by a cross-regulatory feedback loop within cells. The twohypothetical topologies shown (left) would be capable of driving similar,although not identical, cycles of eve, runt and odd expression within cells(right). In Tribolium, the relative expression patterns of Eve protein, runttranscript and odd transcript resemble the predicted expression of model2, rather than model 1 (see supporting information from Choe et al., 2006).Expression predictions assume Boolean regulatory logic and equal timedelays for protein synthesis and protein decay (Clark, 2017). (B) Notchsignalling might indirectly synchronise intracellular oscillations of eve, runtand odd across cells, by acting through hairy. This figure shows ahypothetical regulatory network, which synthesises genetic interactionsdocumented from various different arthropod species (Clark, 2017;Eriksson et al., 2013; Nakao, 2015; Pueyo et al., 2008; Stollewerk et al.,2003). The left half of the network (‘oscillator 1’) would synchroniseoscillations of hairy across neighbouring cells, by coupling hairyexpression to Notch signalling (1). The oscillations of hairy would alsoinfluence the phase of the genetic ring oscillator that forms the right handof the network (‘oscillator 2’), by repressing some of its component genes(2). Cross-regulation between these components (3) would coordinatetheir individual expression patterns, enabling fine-scale regulation ofdownstream genes. (C) Genes such as Wnt, caudal, Dichaete and opahave distinct expression patterns within the SAZ, which correlate withdifferent phases of segment patterning. A, anterior; P, posterior. (Basedon Tribolium data from Clark and Peel, 2018.) Note: Wnt and opa havesegment-polarity patterns in the segmented germband. caudal and/orDichaete stripes (not shown) are seen in the anterior SAZ of somespecies, indicating that the clock feeds back on their expression (Chipmanet al., 2004; Clark and Peel, 2018).

5

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

Surprisingly, in the insects Gryllus, Oncopeltus and Tribolium,the Notch ligand Delta is not expressed in the posterior SAZ(Aranda et al., 2008; Auman et al., 2017; Kainz et al., 2011). EitherNotch signalling acts through a different ligand in these species, or itdoes not directly regulate the clock. Delta also does not appear toplay a segmentation role in the honeybee Apis (a simultaneouslysegmenting species), even though it is expressed in stripes at anappropriate time (Wilson et al., 2010).If a role for Notch signalling in sequential segmentation has

indeed been lost in some insect lineages, it is not clear whatmechanism(s) might synchronise cells instead. One possibility is theToll genes, which are thought to influence intercellular affinity andare expressed dynamically in the SAZ across arthropods (Bentonet al., 2016; Paré et al., 2014). However, they seem only to affectmorphogenetic processes downstream of segment establishment,rather than segment patterning. Another possibility that has beenraised is intercellular communication via Tenascin major (Ten-m)(Hunding and Baumgartner, 2017), a transmembrane protein thatwas erroneously identified as aDrosophila pair-rule factor owing toan opa mutation present on the balancer chromosome of its stock(Zheng et al., 2011). However, mutation/knockdown of Ten-m doesnot affect segmentation in either Drosophila or Tribolium (Choeet al., 2006; Zheng et al., 2011), and Ten-m is expressed periodicallyonly after segment-polarity stripes have formed (Baumgartner et al.,1994; Jin et al., 2019).

The changing regulatory context along the SAZThe segmentation clock oscillates in the posterior SAZ and its phaseis read out in the anterior SAZ. Therefore, the ‘wavefront’ can beloosely identified with the boundary between these regions, whichretracts posteriorly across the embryo over time. The posterior SAZand the anterior SAZ are apparently defined by the differentialexpression of specific regulatory factors (‘timing factors’ in ourterminology), which are expressed dynamically over the course ofaxial elongation, determining where and when segment patterningtakes place (Clark and Peel, 2018). Understanding the mechanisticbasis for the wavefront therefore entails characterising (1) theidentities of these factors, (2) how they regulate segmentation geneexpression, and (3) how they themselves are regulated in theembryo.Many genes are specifically expressed in subregions of the SAZ

(Oberhofer et al., 2014). However, most studies to date have focusedonWnt and caudal, supplemented recently byDichaete/Sox21b andodd-paired (opa)/zic. The expression patterns of these genes arerelatively consistent across species (Fig. 2C). Wnt is expressed in asmall zone around the proctodaeum (Janssen et al., 2010). (We notethat this population of cells appears to be distinct from the SAZproper, and may not contribute to segmental tissue.) In Tribolium,two of its receptors are expressed ubiquitously in the embryo, andone is expressed in the anterior SAZ and in segmental stripes(Beermann et al., 2011). caudal is expressed in the posterior SAZ(Copf et al., 2004; Schulz et al., 1998), and Dichaete is expressed ina similar zone to caudal, but does not overlap with posterior Wnt(Clark and Peel, 2018; Janssen et al., 2018; Paese et al., 2018). Incontrast, opa is expressed in the anterior SAZ, i.e. anterior to orslightly overlapping caudal and Dichaete, and also in segmentalstripes (Clark and Peel, 2018; Green and Akam, 2013; Janssen et al.,2011). Across arthropods, Wnt, caudal and Dichaete are required toestablish and maintain the SAZ (Angelini and Kaufman, 2005b;Bolognesi et al., 2008; Chesebro et al., 2013; Copf et al., 2004;McGregor et al., 2008; Miyawaki et al., 2004; Nakao, 2018; Paeseet al., 2018; Schönauer et al., 2016; Shinmyo et al., 2005). In

Tribolium, opa is required for segmentation, following earlier roles inblastoderm formation and head specification (Clark and Peel, 2018).

Caudal and Dichaete are strong candidates for activating thesegmentation clock, as their expression domains roughly correlatewith the extent of its oscillations, and they positively regulate pair-rule gene expression in Drosophila. Caudal has also been shown tobe necessary for eve and runt expression in Parasteatoda(Schönauer et al., 2016). Opa, on the other hand, may beimportant for reading out the phase of the clock, as it activatessegment polarity genes and regulates late pair-rule gene expressionin Drosophila (Clark and Akam, 2016). Given that all three aretranscription factors, they might regulate segmentation by activatingor repressing specific genes, modulating the regulatory effects ofother transcription factors, or switching expression control betweendifferent enhancers. However, the severity of their knockdownphenotypes in sequentially segmenting species means thatuncovering the details may require precisely targeted functionalperturbations, and probably transgenic reporters.

In sequentially segmenting species, the relative expression patternsof different timing factors remain consistent across development,suggesting that they regulate each other’s expression. Wnt is thoughtto act as a posterior organiser (Chesebro et al., 2013; Oberhofer et al.,2014), and we have hypothesised that regulatory interactions betweencaudal, Dichaete and opa drive their sequential expression over time(Clark and Peel, 2018). In addition, caudal has been found to beactivated by Wnt in diverse arthropods (Beermann et al., 2011;Chesebro et al., 2013; McGregor et al., 2008; Miyawaki et al., 2004),whereas Opa, as a Zic factor, might physically bind the Wnt effectorTCF and modulate its effects on downstream genes (Murgan et al.,2015; Pourebrahim et al., 2011). Therefore, although details arecurrently sketchy, it seems probable that the timing factors areintegrated into a regulatory network that ensures the maintenance ofthe SAZ over time, and also governs its gradual posterior retraction.Given the numerous parallels between posterior development inarthropods and posterior development in other bilaterian phyla, asimilar network might have ancestrally coordinated celldifferentiation during axial extension, and only later been exploitedto regulate segmentation.

In the basic clock-and-wavefront model, the clock stops abruptlywhen it is hit by the wavefront. However, in both arthropodsegmentation and vertebrate somitogenesis, segmentation clockoscillations may resolve into narrowing travelling waves before theystabilise, indicating that the clock winds down relatively gradually.The way in which the oscillation period varies along the SAZ isdescribed phenomenologically by a ‘frequency profile’ (Morelliet al., 2009), and this can vary over developmental time, as well asbetween species. Although the shape of the frequency profile is notpredicted to affect segmentation rate or segment size, modelssuggest that a graded profile might make patterning more robust(El-Sherif et al., 2014; Vroomans et al., 2018).

Wnt signalling perturbations distort the size and proportions of theSAZ (as judged by the expression of caudal), and cause equivalentdistortions to the frequency profile (as judged by the expression ofeve) (El-Sherif et al., 2014). This indicates thatWnt signalling affectsthe dynamics of the segmentation clock, and that its effects might bemediated by SAZ timing factors. However, the mechanism formodulating the oscillation period is not clear. One hypothesisproposes that the clock is quantitatively regulated by a morphogengradient of Caudal (El-Sherif et al., 2014; Zhu et al., 2017), but theeffects of specific timing factors are yet to be disentangled andassessed. Currently, it is unknown whether the period of the clock isindeed explicitly determined by the concentrations of particular

6

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

timing factors (i.e. given control of these levels one could producesustained oscillations of arbitrary period), or whether the slowing ofthe segmentation clock is an inherently transient phenomenon (Verdet al., 2014) inseparable from its temporal transition from anoscillating to a non-oscillating state.

Segment patterning by the pair-rule networkReading out the patternIn the anterior SAZ, each segmentation clock cycle resolves into ananterior-to-posterior array of partially overlapping stripes of pair-rulegene expression. Because the pair-rule genes are expressed in a strictsequence across a clock repeat (e.g. first eve, then runt, then odd),they convey unambiguous phase information to the cells they areexpressed in, which provides significant patterning benefits over asingle-gene oscillator (Fig. 3A). The internal organisation of aparasegment consists of at least three distinct segment-polarity states(Jaynes and Fujioka, 2004; Meinhardt, 1982). Therefore, each pair-

rule gene expression repeat must specify at least three output domainsin species with single-segment periodicity, and at least six outputdomains in species with double-segment periodicity (Fig. 3B).

In Drosophila, the relative expression patterns of pair-rule genesand segment-polarity genes have been characterised in a variety ofgenetic backgrounds, allowing us to infer the regulatory interactionsinvolved in specifying and resolving the segment pattern (reviewedby Clark and Akam, 2016; Jaynes and Fujioka, 2004). Equivalentdata is generally lacking from other arthropod species. However, sofar as we can tell from what does exist (mainly single or doublestains in wild-type embryos) the overall process appears to be fairlyconserved, at least in its broad outline (Auman and Chipman, 2018;Damen et al., 2005; Green and Akam, 2013; Xiang et al., 2017).

First, the primary pair-rule genes pattern the secondary pair-rulegenes. Across arthropods, prd and slp are expressed in a conserved,partially overlapping arrangement, which aligns with prospectiveparasegment boundaries (Choe and Brown, 2007; Green and Akam,

A

B

1 cycle

Palindromic repeat unit

Expression:

Interpretation:

Fate:

Sensitive tonoise

Robust tonoise

Single-gene oscillator

Three-gene oscillator

Expression:

Interpretation:

Fate:

Single-segment periodicity

Parasegmentboundaries

1 cycle

1 parasegment 1 parasegment

Polarised repeat unit

Input:

Output: A P A P AA PA PAP P

C Dynamic model for the patterning of prd and slp

Eve

slpprd

t1t2

PosteriorAnterior

Patternedearlier

Patternedlater

D Segment-polarity genes are patterned by a ‘combinatorial code’ of pair-rule gene expression

PosteriorAnterior

Pair-rule repeat

EveRunt

FtzOdd

PrdSlp

en

Eve

FtzOdd

PrdSlp

Inputs:

Output:

Odd-numbered

stripe

Control logic: Even-numbered stripesOdd-numbered stripes

en

Prd Slp RuntOdd

en

Ftz Slp Odd

Eve stripeshifts anteriorly

over time

Even-numbered

stripe

Odd-numbered

stripe

Parasegmentboundaries

Double-segment periodicity

D E F A B C D E F A B C D E F A B C

C B A A B C C B A A B C C B A A B C

enen

AA AAPP PP

Fig. 3. Resolving the segment pattern: fromoscillations to stable stripes. (A) Comparison of patterning using a single-gene oscillator versus patterning usinga three-gene oscillator. With a single-gene oscillator, different cell fates are determined by different expression levels of the oscillator. The output is sensitive tonoise in the amplitude of, or measuring of, the signal, and must be palindromic, because the input signal is symmetrical. With a three-gene oscillator, different cellfates can be determined by different combinations of input factors. The output is more robust to noise, and has an inherent polarity. (B) Comparison of thesegment-polarity fate readout for three-gene oscillator clocks with single-segment or double-segment periodicity. Parasegment boundaries (red lines) formwherever a cell with an anterior segment-polarity fate (‘A’; i.e. expressing engrailed) abuts a cell with a posterior segment-polarity fate (‘P’; i.e. expressing slpand wg). A third cell fate (light grey; e.g. odd in Drosophila) prevents ectopic boundaries. Note that species with double-segment periodicity have a different,more complex mapping between the input pattern (pair-rule gene expression) and the output pattern (segment-polarity gene expression). (C) Dynamicmodel for the patterning of prd and slp in Drosophila: the staggered expression boundaries of prd and slp are caused by the Eve stripes shifting anteriorlyacross the tissue over time. The posterior border of the prd stripe is patterned at time point t1 (Eve expression shown by dashed line); the posterior border ofthe slp stripe is patterned a short while later, at time point t2 (Eve expression shown by solid line). (Based on Clark, 2017.) (D) The staggered pattern of pair-rulegene expression comprises a positional code, which specifies narrow stripes of segment-polarity gene expression. The regulatory logic (top) and resultingexpression pattern (bottom) of Drosophila engrailed (en) is shown as an example. Note that the input pattern has double-segment periodicity, andodd-numbered and even-numbered en stripes are regulated differently. (Based on Jaynes and Fujioka, 2004.)

7

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

2013). In both Drosophila and other arthropods, prd turns on earlierthan slp, at a time when upstream pair-rule gene expression is stilldynamic. In Drosophila, both genes are patterned by Eve, and wehave proposed that the dynamic nature of the Eve stripes (see below)helps differentially position the two domains (Clark, 2017) (Fig. 3C).Next, the segment-polarity genes are activated. Each segment-

polarity gene is activated or repressed by particular pair-rule factors,which combinatorially define where it is expressed within thepattern repeat (Bouchard et al., 2000; Choe and Brown, 2009;DiNardo and O’Farrell, 1987). In species with double-segmentperiodicity, odd-numbered and even-numbered segment-polaritystripes may be driven by different regulatory logic (Fig. 3D).At the same time, some of the pair-rule genes also start being

expressed in segment-polarity patterns. In pair-rule species, thisinvolves the splitting of existing stripes or the intercalation of newones. The new patterns are explained by a new network of regulatoryinteractions between the pair-rule genes (Clark and Akam, 2016). Incontrast to the earlier network, which drives dynamic expression,this later one behaves like a multistable switch, ‘locking in’ specificsegment-polarity fates (Clark, 2017). Interestingly, differentprimary pair-rule genes undergo frequency doubling in each ofDrosophila, Bombyx, Tribolium and Nasonia (Choe et al., 2006;Clark and Akam, 2016; Nakao, 2015; Rosenberg et al., 2014),contrasting with the conserved expression of the segment-polarityand secondary pair-rule genes.The resulting segmental patterns go on to regulate morphological

segmentation. Note that the pair-rule genes are therefore pleiotropic:they are involved in generating the segment pattern, but someadditionally play roles in maintaining segment polarity, and they alsoregulate the development of other structures, such as the nervoussystem. In some cases, these functions have become distributedbetween multiple paralogues, e.g. prd/gooseberry/gooseberry-neuroin Drosophila (He and Noll, 2013), or the three copies of eve inStrigamia (Green and Akam, 2013). Across species, there can beconsiderable variation in both the number of paralogues present in thegenome and the degree of subfunctionalisation between them,complicating the interpretation of genetic perturbations.

The evolution of pair-rule patterningIn several insect species, and also the centipede Strigamia(Chipman et al., 2004), segmentation gene expression undergoesa striking transition from double-segment periodicity to single-segment periodicity as the segment pattern is resolved. However,there is no indication of an initial double-segment periodicity duringsequential segmentation in the spiders Cupiennius (Davis et al.,2005; Schoppmeier and Damen, 2005a) and Parasteatoda(Schwager, 2008), the millipede Glomeris (Janssen et al., 2011),or the crustacean Daphnia (Eriksson et al., 2013) (Fig. 1A). Thissuggests that the ancestral arthropod segmentation clock had asingle-segment periodicity, and that pair-rule patterning in insectsand centipedes originated independently.Beyond this, it is not clear exactly when or how many times pair-

rule patterning evolved in either of the centipede or insect lineages.eve is expressed segmentally rather than in pair-rule stripes in adifferent centipede species, Lithobius (Hughes and Kaufman,2002b), which could indicate that pair-rule patterning evolvedrelatively recently within the centipede clade, possibly correlatingwith the origin of longer-bodied forms. However, the dynamics ofthe Lithobius segmentation clock will need be investigated to ruleout a transient or cryptic double-segment periodicity.In insects, most of the available data come from holometabolan or

orthopteran species, as well as the cockroach Periplaneta and

the hemipteran bug Oncopeltus (Fig. 1A). Holometabolans (Binnerand Sander, 1997; Nakao, 2010; Patel et al., 1994; Rosenberg et al.,2014) and orthopterans (Davis et al., 2001; Mito et al., 2007) bothshow obvious transitions from double-segment to single-segmentperiodicity, but the mapping between the pair-rule pattern and thesegmental pattern is different in the two groups, suggesting that theirrespective pair-rule mechanisms might have evolved independently.Consistent with this possibility, gene expression in Periplaneta(more closely related to orthopterans than to holometabolans)appears to be single-segmental (Pueyo et al., 2008), although, aswith Lithobius, the dynamics of its segmentation clock have notbeen explicitly investigated. Finally, Oncopeltus is a rather strangecase: based on the expression and function of eve, it appears to lackpair-rule patterning, but pair-rule expression and/or function ofcertain other genes hints at an underlying double-segmentperiodicity (Auman and Chipman, 2018; Benton et al., 2016;Erezyilmaz et al., 2009; Liu and Kaufman, 2005; Reding et al.,2019).

Thus, although the evidence from some of these species isambiguous, the current picture suggests that pair-rule patterningmay have evolved within crown-group insects, possibly multipletimes. This is puzzling, because the specialised and relativelyinvariant body plan of insects presents a morphological constraintthat is hard to reconcile with a saltational doubling of segmentationrate. [Instead, it is much easier to imagine pair-rule patterningevolving in remipedes, which are thought to be the sister group ofhexapods (Schwentner et al., 2017), and have homonomous,centipede-like bodies.] How was the evolution of double-segmentperiodicity coordinated with compensatory changes to Hoxdynamics and the duration of axial extension, in order to keepsegment number (Box 2) and segment identity constant? Given thatStrigamia seems to switch to a single-segment periodicity whenadding its most posterior segments (Brena and Akam, 2013), andthat pair-rule patterns are seen during the anterior patterning ofotherwise segmental species (Dearden et al., 2002; Janssen et al.,2012), one possibility is that pair-rule patterning was introducedgradually along the AP axis, allowing other developmentalparameters the chance to adapt.

As pair-rule patterning requires half the number of clock cycles togenerate a given number of segment-polarity stripes, its evolutionmay have been driven by selection for faster development (inholometabolans) or a longer body (in centipedes). However, it iscurrently not obvious how the ancestral segment-patterningmechanism was modified to become pair-rule. Segmentalfrequency could have been doubled by changing the ‘readout’ ofa conserved clock, i.e. by evolving new enhancers to driveadditional segment-polarity stripes in between the originals, oraltering the control logic of existing enhancers to drive a pair ofstripes instead of just one. Alternatively, the clock itself couldhave been modified, e.g. by recruiting new genes into the originalcyclic repeat and thereby expanding its patterning potential. Toreconstruct the specific regulatory changes that occurred, it will beinformative to find out how the gene expression and enhancerlogic of pair-rule species compares with their closest segmentalrelatives.

The evolution of simultaneous segmentationReconciling sequential and simultaneous segmentationA segmentation clock is one strategy for generating periodicity, butanother is simply to regulate each stripe individually, exploitingwhatever positional information is locally available (François et al.,2007; Salazar-Ciudad et al., 2001; Vroomans et al., 2016). This

8

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

latter method is used in the Drosophila blastoderm, where over 20‘stripe-specific elements’ (SSEs) regulate the expression of the fiveprimary pair-rule genes (Schroeder et al., 2011). These elementsreceive spatial information from gap factors, and each drivesexpression at a different AP position (or pair of positions) alongthe blastoderm, contributing just one or two stripes to a gene’soverall 7-stripe pattern. Sepsid flies (which diverged fromdrosophilids about 100 million years ago) are also known to usethis kind of element (Hare et al., 2008), and it is likely thatsimilarly ad hoc regulatory mechanisms are used whereverperiodicity emerges simultaneously, e.g. in the blastoderms ofNasonia (Rosenberg et al., 2014) and Oncopeltus (Stahi andChipman, 2016), or in the chelicerate prosoma (Pechmann et al.,2011; Schwager et al., 2009). Although less ‘elegant’ than usingtemporal oscillations, this explicitly spatial mode of segmentationcan, in principle, occur much faster, because a number of differentpattern repeats can be initialised at once.Simultaneous segmentation, typified by Drosophila, is

traditionally thought of as mechanistically distinct from sequentialsegmentation, typified by, for example, Tribolium or Gryllus. Thetextbook model of the hierarchical ‘subdivision’ of a syncytialblastoderm by morphogen gradients seems a world away fromwaves of gene expression within a cellularised, elongatinggermband. However, the Drosophila blastoderm is now known tobe more dynamic than was previously imagined, and the basicstructure of its segment patterning network seems remarkablysimilar to that of other arthropods (Fig. 4A).

As the Drosophila blastoderm stage is so short, the effects ofdynamic gene expression are subtle, and for years were overlooked.However, quantitative expression atlases suggest that expressiondomains in the posterior half of the blastoderm travel anteriorlyacross cells over time (Jaeger et al., 2004; Keränen et al., 2006;Surkova et al., 2008), and this has recently been demonstratedthrough live imaging (El-Sherif and Levine, 2016; Lim et al., 2018).The shifts reflect sequential patterns of transcriptional states withincells, and trace back to asymmetric repressive interactions in the gapgene network (Jaeger, 2011; Verd et al., 2018) (Fig. 4Bi) – perhapssimilar to those driving their temporal expression in the SAZs ofsequentially segmenting species.

In theDrosophila blastoderm, the expression dynamics of the gapgenes are directly transferred to pair-rule genes via their SSEs(Fig. 4Bii). In addition, the pair-rule genes cross-regulate each otherthrough ‘zebra elements’: enhancers that drive expression in all ofthe trunk stripes simultaneously (Schroeder et al., 2011). (Someprimary pair-rule genes, and both secondary pair-rule genes,possess zebra elements.) These regulatory interactions are alsodynamic, and they combine with the stripe shifts driven by the gapgenes to generate a staggered sequence of pair-rule gene expressionwithin each double-segment repeat (Clark, 2017) (Fig. 4Biii). Thisspatiotemporal sequence is the same as that driven by thesegmentation clock in sequentially segmenting species such asTribolium and Strigamia (Choe et al., 2006; Green and Akam,2013), suggesting that zebra enhancers and ‘clock’ enhancers maybe homologous.

Once primary pair-rule gene expression is properly phased withineach double-segment repeat, Drosophila segment patterningproceeds just as it would in the anterior SAZ of a sequentiallysegmenting species, beginning with the activation of prd and slp,and moving on to segment-polarity gene expression and stripedoubling. This conserved process of pattern resolution is apparentlyregulated by a conserved sequence of timing factor expression:posterior SAZ factors Caudal and Dichaete are expressedthroughout the trunk during the early, dynamic stages of pair-rulegene expression inDrosophila, and are replaced by the anterior SAZfactor Opa as the segment-polarity pattern is being resolved (Clarkand Peel, 2018).

The Drosophila blastoderm therefore seems effectivelyequivalent to a SAZ, except that rather than maturing graduallyfrom anterior to posterior, it does so all at once (Fig. 4C).We suspectthat much of the ancestral segmentation machinery remains intact.However, as spatial information is no longer conveyed by thedelayed maturation of posterior tissue, gap genes and SSEs preloadit into the system instead (Fig. 4A). Importantly, although geneticperturbations tend to result in different phenotypes in the two modesof segmentation (e.g. primary pair-rule genes cause pair-rulephenotypes in Drosophila rather than truncations), this mightoften be explained by the divergent deployment of the genes in theembryo, rather than divergent function.

The evolution of stripe-specific elementsSimultaneous segmentation differs from sequential segmentation intwo key respects: its temporal regulation (determined by theexpression profiles of the timing factors), and the spatial pre-patterning of the pair-rule genes by gap genes (Fig. 4C).Simultaneous segmentation is also associated with an anteriorshift of the blastoderm fate map and an increase in the number ofsegments patterned prior to gastrulation. [Note, however, thatalthough segment patterning in the blastoderm is often simultaneousand regulated by gap genes, this need not be the case: Tribolium

Box 2. Regulation of segment numberIn arthropods, segment number is determined by the total number of pair-rule stripes (and the periodicity with which they regulate segment-polaritygenes). In simultaneously segmenting insects, such as Drosophila,individual pair-rule stripes are positioned by gap factors at specificlocations along the AP axis, hardcoding segment number. Insequentially segmenting species, segment number instead dependson the temporal duration of segmentation, divided by the period of thesegmentation clock.

Gap genes appear to play some role in controlling the duration ofsegment addition (Cerny et al., 2005; Nakao, 2016). Over time, gapgenes are expressed sequentially within the SAZ, their turnover drivenby cross-regulatory interactions (Boos et al., 2018; Verd et al., 2018).This process, effectively a developmental ‘timer’, shows intriguingsimilarities to the ‘neuroblast clock’ (Isshiki et al., 2001; Peel et al.,2005). It evidently exerts some control over the body plan, as perturbinghunchback expression can both decrease (Liu and Kaufman, 2004;Marques-Souza et al., 2008; Mito et al., 2005) and increase (Boos et al.,2018; Nakao, 2016) segment number in sequentially segmentinginsects. These phenotypes are not well understood, but might resultfrom gap genes directly or indirectly regulating cell behaviour within theSAZ. Such effects are unlikely to be mediated via the Hox genes,because significant perturbations of Hox gene expression in insects andcrustaceans have not been found to affect segment number (Angeliniet al., 2005; Martin et al., 2016; Stuart et al., 1991).

Despite varying widely among arthropods, segment number is usuallyfixed within a species. However, there are certain groups, such asgeophilomorph centipedes, in which naturally occurring variation mightprovide clues as to how this number evolves (Kettle and Arthur, 2000;Vedel et al., 2008, 2010). Another interesting question is how speciesthat undergo post-embryonic segmentation coordinate segmentpatterning with the moult cycle. Ecdysone-related genes playsegmentation roles in some embryos (Erezyilmaz et al., 2009; Hefferet al., 2013), suggesting that these two processes might be deeplyrelated.

9

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

A Sequential segmentatione.g. Tribolium

Gap genenetwork

Primarypair-rule gene

network

Segment-polarity/late pair-rule

network

Timingfactor

network(wavefront)

Secondarypair-rule genes

B

A P

even skippedengrailed

Kr

Kni

Gt

(i) Gap gene network cross-regulationcauses expression shifts

(ii) Stripes of SSE-driven pair-rule genes shift anteriorly

(iii) Shifts plus cross-repressionorganise pair-rule pattern

eve5 4+6

Kr Gt Kni Hb

oddzebra

Eve Hairy

runtzebra

OddHairy

Unstable but functionallyimportant overlaps

Fate specification Spatiotemporalregulation

Duration/termination

?

Posterior

Anterior

Gap genenetwork

Fate specification

Simultaneous segmentatione.g. Drosophila

Anteriorpatterning

centre (spatial)

Spatiotemporalregulation

Time

A P

Time

Sequential segmentation

Simultaneous segmentation

C

even skippedengrailed

Primarypair-rule gene

network

Segment-polarity/late pair-rule

network

Timingfactor

network(temporal)

Secondarypair-rule genes

Earlier

Later

Wavefront

SSEs

Fig. 4. Reconciling sequential andsimultaneous segmentation.(A) Structural overview of arthropodsegmentation gene networks. The coreof the system (yellow box) is relativelyconserved across species. In sequentialsegmentation, spatial information isprovided by the timing factor network,which generates a wavefront. Gap genesdo not play a major role in segmentpatterning, although late gap geneexpression may be important forterminating segmentation, by repressingtiming factors that maintain the SAZ(dashed blue line). In simultaneoussegmentation, timing factors only providetemporal information. Spatial informationis usually provided by a novel anteriorpatterning centre (i.e. a morphogengradient such as Bicoid; Liu et al., 2018;McGregor, 2005), which regulates gapgene expression. Gap genes pass thisinformation to the primary pair-rulegenes, through newly evolved regulatoryelements (SSEs). (B) Spatial patterningin Drosophila is inherently dynamic. (i)Regulatory interactions between gapgenes cause gap domains to shiftanteriorly across the blastoderm overtime. (ii) Stripes of pair-rule geneexpression regulated by gap inputs alsoshift anteriorly. (iii) Regulatoryinteractions between the pair-rule genesconvert these shifts into a staggeredpattern of expression overlaps across thepair-rule repeat. Note that each panelzooms in on a smaller region of the APaxis. (C) Schematic kymographs (i.e.plots of how gene expression along theAP axis changes over time) comparingthe key spatiotemporal features ofsequential and simultaneoussegmentation. In sequentialsegmentation, timing factor expression(blue) matures from anterior to posterioracross the tissue, producing a wavefront(diagonal line). Periodicity is generatedby sustained oscillations (note how evenskipped turns on and off over time withinthe blue zone). The wavefront convertsthe oscillations into a stable segment-polarity pattern (engrailed expression). Insimultaneous segmentation, there is littlespatial regulation of timing factorexpression across the tissue, and pair-rule stripes are present from the start.Embryo diagrams depict the specific timepoints they line up with on thekymographs (eve expression is notshown). Patterning has double-segmentperiodicity. Note that the two time axeshave different scales.

10

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

A Each SSE can take over from the clock gradually

(i) Along the AP axis

(ii) Within each clock repeat

Clock enhancer more important SSE more important

SSE 1 clockSSE 2SSE 3SSE 1 clockSSE 2SSE 1 clockclock

More segments generated by the clock Fewer segments generated by the clock

1 repeat

More stripes established by cross-regulationpatterning is more dynamic

More stripes established by gap genespatterning can be less dynamic

Magnitude ofstripe shift

Driven by clockDriven by SSE

Evolution ofnew SSE

Segments patternedby new SSE

Segments patternedby the clock

SSE clock

Gapgenes

Pair-rulegenes

SSE

Gapgenes

eve 54+6

Kr GtKni Hb

3+7

Kni Hb

eve 54+6

Kr GtKni Hb

3

Kni Hb

eve 54

Kr GtKni Hb

3

Kni Hb

3 4 5 3 4 5 6 3 4 5 6 7

Duplicatedstripe 4

Duplicatedstripe 3

Simpler gap gene pattern More complex gap gene pattern

Segments patternedby new SSE

Segments patternedby new SSE

Evolution ofnew SSE

Evolution ofnew SSE

SSE clock

Gapgenes

Pair-rulegenes

clock

Pair-rulegenes

A P

Enhanceractivity

Novel Hbdomain

B SSEs are able to evolve one at a time

C Existing SSEs can be recruited to drive additional stripes

(i) Stripe driven by clock enhancer

(ii) SSE shadows clock enhancer

(iii) SSE establishes stripe; clock enhancer then refines it

(iv) Once all stripes driven by SSEs, clock enhancer may be lost

Fig. 5. The evolution of simultaneous segmentation involves a gradual replacement of the segmentation clock by SSEs. (A) Clock enhancers(potentially homologous to zebra elements) and SSEs both drive stripes that shift anteriorly over time. SSEs can therefore gradually assume regulatorycontrol over particular clock-driven stripes (i-iv), without disrupting downstream patterning. (B) Simultaneous patterning is likely to evolve stepwise along theAP axis, via the acquisition over evolutionary time of new SSEs that control expression in increasingly posterior stripes. Embryo diagrams assume asegmentation clock with double-segment periodicity. In addition, simultaneous patterning is likely to evolve stepwise within each pair-rule gene expressionrepeat, as more of the primary pair-rule genes evolve their own SSEs. Additional SSEs reduce the time required to organise pair-rule gene expressionacross the repeat. As a consequence, the magnitude of the stripe shifts can decrease. (C) Changes in gap gene expression can be sufficient to generateadditional SSE-driven stripes, without accompanying changes in cis-regulatory logic. InDrosophila (right), SSEs such as eve 3+7 and eve 4+6 each drive a pair ofstripes. The current situation likely evolved from a simpler scenario (left), in which the same enhancers drive expression in only one stripe each. Gt, Giant; Hb,Hunchback; Kni, Knirps; Kr, Kruppel. Note that eve 3+7 and eve 4+6 are both repressed by Kni and Hb, but with different relative strengths, represented bydifferent arrow thicknesses (Samee et al., 2017). Diagrams are colour-coded such that transcription factor names (top) have the same colour as theircorresponding expression domain(s) (below).

11

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

patterns its blastoderm segments sequentially, using retractingtiming factors and a clock (El-Sherif et al., 2014, 2012)].The evolution of simultaneous segmentation appears to be

constrained by early embryogenesis (French, 1988). Some insects,such as orthopterans, have ‘panoistic’ ovaries, in which all germlinecells become oocytes, and the eggs contain little but yolk (Büning,1994). These species pattern their segments sequentially. Otherinsects, such as hemipterans and holometabolans, have ‘meroistic’ovaries, in which germline-derived ‘nurse’ cells load oocytes withmaternal mRNA. These species frequently have a biphasic mode ofsegmentation, in which anterior segments are patternedsimultaneously. Meroistic ovaries (which facilitate pre-patterningof the egg), may therefore be a pre-adaptation for simultaneoussegmentation.Extreme examples of simultaneous segmentation (e.g.

Drosophila) have evolved independently within each of the majorholometabolan orders (Davis and Patel, 2002). [Intriguingly, therehas also been at least one reversion to sequential segmentation,within braconid wasps (Sucena et al., 2014)]. A Drosophila-likemode of segmentation likely requires far-reaching changes to earlyembryogenesis, such as a novel anterior patterning centre to helpspatially pattern gap genes along the entire AP axis of the egg(Lynch et al., 2006) (Fig. 4A). Here, we focus on understanding howSSEs and gap genes are together able to take over stripe patterningfrom the clock. It seems likely that this transition to intricate spatialregulation involves a series of selectively favourable regulatorychanges, which incrementally increase the speed or robustness ofsegmentation, while strictly preserving its output (Fig. 5).First, new SSEs seem to be easy to evolve, because they tend to be

short, with simple regulatory logic and high sequence turnoverbetween closely related species (Hare et al., 2008; Ludwig et al.,1998). Some of them may have been selected simply to increase therobustness of segmentation clock expression; this might haveoccurred in either a blastoderm or a SAZ context. [There is onereport from Tribolium suggesting the existence of SSEs thatdrive expression in the germband (Eckert et al., 2004)]. Importantly,because gap gene expression is inherently dynamic (whether in theblastoderm or the SAZ), SSE-regulated stripes are predicted to‘shadow’ stripes driven by the clock, allowing them to take overdownstream functions gradually (Verd et al., 2018) (Fig. 5A).Second, only a single new SSE need evolve at one time.

Simultaneous patterning seems likely to have evolved progressively,from anterior to posterior, with each new SSE-driven stripe reducingthe number of cycles needed from the clock (Peel and Akam, 2003)(Fig. 5Bi). Furthermore, cross-regulation between the pair-rule genesmeans that an SSE for one gene could in principle go on to organise awhole pattern repeat, with the remaining genes evolving their ownSSEs afterwards, to make patterning faster or more robust (Clark,2017) (Fig. 5Bii). This process might be highly contingent: inDrosophila, eve and runt have full sets of SSEs and odd is patternedlargely through cross-regulation (Schroeder et al., 2011), but RNAievidence fromBombyx suggests precisely the opposite (Nakao, 2015).Finally, SSEs can be reused. InDrosophila there are several SSEs

that drive a pair of stripes, typically arranged symmetrically arounda particular gap domain (Schroeder et al., 2011). This suggests thatposterior gap gene expression evolved to duplicate the regulatoryenvironments of anterior stripes, thereby initialising additional pair-rule gene stripes without the need to evolve additional SSEs(Fig. 5C).Interestingly, Drosophila eve stripes 3 and 7, which are co-driven

by a single SSE, are regulated by the same gap genes as are evestripes 3 and 6 in Anopheles (Goltsev et al., 2004), which has led to a

proposal that certain stripes have been lost or gained from theselineages over time (Rothschild et al., 2016). This hypothesis is hardto reconcile with the gradualist scenariowe favour, as the transitionalstates would have severely compromised fitness. We think it morelikely that the posterior gap gene domains were recruited in adifferent order in the Drosophila and Anopheles lineages, resultingin a homologous ‘stripe 3’ element additionally driving non-homologous posterior stripes. In support of this alternative, a midgespecies more closely related to Drosophila than to Anophelespatterns only five eve stripes before gastrulation (Rohr et al., 1999),indicating that Anopheles and Drosophila probably evolved fullysimultaneous segmentation independently (Jaeger, 2011).

ConclusionsOur current understanding is that arthropod segment patterning is aninherently dynamic and a significantly conserved process, ancestrallytaking the form of a clock-and-wavefront system. Note, however, thatmany of the conclusions in this Review extrapolate from fragmentarydata gathered from a small number of model species, with functionaldata available from an even smaller number. This is certainly not thelast word on arthropod segmentation, but we hope to have provided acoherent framework for further thought and experiment.

We anticipate that future investigation will centre on twocontrasting but inter-related tasks. First, better resolving the natureof the ancestral arthropod clock-and-wavefront system: the topologyof the gene regulatory networks comprising the clock, the productionof timing factor wavefronts by a retracting SAZ, and the mechanisticbasis for the interactions between them. Second, reconstructing howarthropod segmentation networks have diversified over time, givingrise to such remarkable novelties as simultaneous patterning anddouble-segment periodicity. In addition, we believe that sequentiallysegmenting arthropod models are well placed to complement andinform the study of vertebrate axial patterning, especially given theirbenefits of cost-efficiency, short generation times, experimentaltractability, and relatively simple genomes.

The most pressing next step is to collect good-quality multiplexedexpression data from a variety of arthropod species (Choi et al., 2018,2016) and cross-reference this with information about tissue dynamics(Wolff et al., 2018), to better characterise how segmentation geneexpression changes over space and time. Building on a soliddescriptive foundation, there are numerous exciting directions topursue: genome editing to generatemutants,misexpression constructs,and live reporters (Gilles et al., 2015; Lai et al., 2018); constructionand analysis of data-informed dynamical models (Sharpe, 2017);single-cell sequencing of segmenting tissues (Griffiths et al., 2018); exvivo culturing of SAZ cells (Lauschke et al., 2013). Over the past fourdecades, arthropod segmentation has contributed enormously to ourunderstanding of developmental gene networks and their evolution.As we enter a new ‘golden age’ of developmental biology, we seegreat promise for this legacy to continue.

AcknowledgementsWe thank Olivia Tidswell, Matthew Benton, Lauren Bush, Mariana Wolfner andRoger Keynes for comments on the manuscript.

Competing interestsThe authors declare no competing or financial interests.

FundingThe authors were supported by Biotechnology and Biological Sciences ResearchCouncil research grants (BB/P009336/1 to M.A. and E.C.; BB/L020092/1 to A.D.P.).E.C. was also supported by a Junior Research Fellowship from Trinity College,University of Cambridge, and a European Molecular Biology Organization LongTerm Fellowship.

12

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

ReferencesAkam, M. (1987). The molecular basis for metameric pattern in the Drosophilaembryo. Development 101, 1-22.

Anderson, D. T. (1973). Embryology and Phylogeny in Annelids and Arthropods.,International series of Monographs in Pure and Applied Biology. Oxford:Pergamon Press.

Angelini, D. R. and Kaufman, T. C. (2005a). Insect appendages and comparativeontogenetics. Dev. Biol. 286, 57-77. doi:10.1016/j.ydbio.2005.07.006

Angelini, D. R. and Kaufman, T. C. (2005b). Functional analyses in the milkweedbug Oncopeltus fasciatus (Hemiptera) support a role for Wnt signaling in bodysegmentation but not appendage development. Dev. Biol. 283, 409-423. doi:10.1016/j.ydbio.2005.04.034

Angelini, D. R., Liu, P. Z., Hughes, C. L. and Kaufman, T. C. (2005). Hox genefunction and interaction in the milkweed bug Oncopeltus fasciatus (Hemiptera).Dev. Biol. 287, 440-455. doi:10.1016/j.ydbio.2005.08.010

Aranda, M., Marques-Souza, H., Bayer, T. and Tautz, D. (2008). The role of thesegmentation gene hairy in Tribolium. Dev. Genes Evol. 218, 465-477. doi:10.1007/s00427-008-0240-1

Auman, T. and Chipman, A. D. (2018). Growth zone segmentation in the milkweedbug Oncopeltus fasciatus sheds light on the evolution of insect segmentation.BMC Evol. Biol. 18, 178. doi:10.1186/s12862-018-1293-z

Auman, T., Vreede, B. M. I., Weiss, A., Hester, S. D., Williams, T. A., Nagy, L. M.and Chipman, A. D. (2017). Dynamics of growth zone patterning in the milkweedbug Oncopeltus fasciatus.Development 144, 1896-1905. doi:10.1242/dev.142091

Azpiazu, N., Lawrence, P. A., Vincent, J. P. and Frasch, M. (1996). Segmentationand specification of the Drosophila mesoderm. Genes Dev. 10, 3183-3194.doi:10.1101/gad.10.24.3183

Baumgartner, S., Martin, D., Hagios, C. and Chiquet-Ehrismann, R. (1994).Tenm, a Drosophila gene related to tenascin, is a new pair-rule gene. EMBO J. 13,3728-3740. doi:10.1002/j.1460-2075.1994.tb06682.x

Beermann, A., Pruhs, R., Lutz, R. and Schroder, R. (2011). A context-dependentcombination of Wnt receptors controls axis elongation and leg development in ashort germ insect. Development 138, 2793-2805. doi:10.1242/dev.063644

Benton, M. A. (2018). A revised understanding of Tribolium morphogenesis furtherreconciles short and long germ development. PLoS Biol. 16, e2005093. doi:10.1371/journal.pbio.2005093

Benton, M. A., Pechmann, M., Frey, N., Stappert, D., Conrads, K. H., Chen, Y.-T.,Stamataki, E., Pavlopoulos, A. and Roth, S. (2016). Toll genes have anancestral role in axis elongation. Curr. Biol. 26, 1609-1615. doi:10.1016/j.cub.2016.04.055

Binner, P. and Sander, K. (1997). Pair-rule patterning in the honeybee Apismellifera: expression of even-skipped combines traits known from beetles andfruitfly. Dev. Genes Evol. 206, 447-454. doi:10.1007/s004270050074

Blower, J. G. (1985). Millipedes, Linnean Society Synopses of the British Fauna.London: E.J. Brill/Dr. W. Backhuys.

Bolognesi, R., Farzana, L., Fischer, T. D. and Brown, S. J. (2008). Multiple Wntgenes are required for segmentation in the short-germ embryo of Triboliumcastaneum. Curr. Biol. 18, 1624-1629. doi:10.1016/j.cub.2008.09.057

Boos, A., Distler, J., Rudolf, H., Klingler, M. and El-Sherif, E. (2018). A re-inducible gap gene cascade patterns the anterior–posterior axis of insects in athreshold-free fashion. eLife 7, e41208. doi:10.7554/eLife.41208

Bouchard, M., St-Amand, J. and Cote, S. (2000). Combinatorial activity of pair-ruleproteins on the Drosophila gooseberry early enhancer. Dev. Biol. 222, 135-146.doi:10.1006/dbio.2000.9702

Brena, C. and Akam, M. (2012). The embryonic development of the centipedeStrigamia maritima. Dev. Biol. 363, 290-307. doi:10.1016/j.ydbio.2011.11.006

Brena, C. and Akam, M. (2013). An analysis of segmentation dynamics throughoutembryogenesis in the centipede Strigamia maritima. BMC Biol. 11, 112. doi:10.1186/1741-7007-11-112

Brusca, R. C., Moore, W. and Shuster, S. M. (2016). Invertebrates, 3rd edn.Sunderland, MA: Sinauer.

Buning, J. (1994). The Insect Ovary: Ultrastructure, Previtellogenic Growth andEvolution. Netherlands: Springer.

Cepeda, R. E., Pardo, R. V., Macaya, C. C. and Sarrazin, A. F. (2017). Contributionof cell proliferation to axial elongation in the red flour beetle Tribolium castaneum.PLoS ONE 12, e0186159. doi:10.1371/journal.pone.0186159

Cerny, A. C., Bucher, G., Schroder, R. and Klingler, M. (2005). Breakdown ofabdominal patterning in the Tribolium Kruppel mutant jaws. Development 132,5353-5363. doi:10.1242/dev.02154

Chesebro, J. E., Pueyo, J. I. and Couso, J. P. (2013). Interplay between a Wnt-dependent organiser and the Notch segmentation clock regulates posteriordevelopment in Periplaneta americana. Biol. Open 2, 227-237. doi:10.1242/bio.20123699

Chipman, A. D. (2010). Parallel evolution of segmentation by co-option of ancestralgene regulatory networks. BioEssays 32, 60-70. doi:10.1002/bies.200900130

Chipman, A. D. and Akam, M. (2008). The segmentation cascade in the centipedeStrigamia maritima: Involvement of the Notch pathway and pair-rule genehomologues. Dev. Biol. 319, 160-169. doi:10.1016/j.ydbio.2008.02.038

Chipman, A. D., Arthur, W. and Akam, M. (2004). A double segment periodicityunderlies segment generation in centipede development. Curr. Biol. 14,1250-1255. doi:10.1016/j.cub.2004.07.026

Choe, C. P. and Brown, S. J. (2007). Evolutionary flexibility of pair-rule patterningrevealed by functional analysis of secondary pair-rule genes, paired and sloppy-paired in the short-germ insect, Tribolium castaneum. Dev. Biol. 302, 281-294.doi:10.1016/j.ydbio.2006.09.037

Choe, C. P. andBrown, S. J. (2009). Genetic regulation of engrailed andwingless inTribolium segmentation and the evolution of pair-rule segmentation. Dev. Biol.325, 482-491. doi:10.1016/j.ydbio.2008.10.037

Choe, C. P., Miller, S. C. and Brown, S. J. (2006). A pair-rule gene circuit definessegments sequentially in the short-germ insect Tribolium castaneum. Proc. Natl.Acad. Sci. USA 103, 6560-6564. doi:10.1073/pnas.0510440103

Choi, H. M. T., Calvert, C. R., Husain, N., Huss, D., Barsi, J. C., Deverman, B. E.,Hunter, R. C., Kato, M., Lee, S. M., Abelin, A. C. T. et al. (2016). Mapping amultiplexed zoo of mRNA expression. Development 143, 3632-3637. doi:10.1242/dev.140137

Choi, H. M. T., Schwarzkopf, M., Fornace, M. E., Acharya, A., Artavanis, G.,Stegmaier, J., Cunha, A. and Pierce, N. A. (2018). Third-generation in situhybridization chain reaction: multiplexed, quantitative, sensitive, versatile, robust.Development 145, dev165753. doi:10.1242/dev.165753

Clark, R. B. (1964).Dynamics in Metazoan Evolution: The Origin of the Coelom andSegments. Clarendon Press.

Clark, E. (2017). Dynamic patterning by the Drosophila pair-rule network reconcileslong-germ and short-germ segmentation. PLoS Biol. 15, e2002439. doi:10.1371/journal.pbio.2002439

Clark, E. and Akam, M. (2016). Odd-paired controls frequency doubling inDrosophila segmentation by altering the pair-rule gene regulatory network. eLife5, e18215. doi:10.7554/eLife.18215

Clark, E. and Peel, A. D. (2018). Evidence for the temporal regulation of insectsegmentation by a conserved sequence of transcription factors. Development145, dev155580. doi:10.1242/dev.155580

Cooke, J. and Zeeman, E. C. (1976). A clock and wavefront model for control of thenumber of repeated structures during animal morphogenesis. J. Theor. Biol. 58,455-476. doi:10.1016/S0022-5193(76)80131-2

Copf, T., Schroder, R. and Averof, M. (2004). Ancestral role of caudal genes in axiselongation and segmentation. Proc. Natl. Acad. Sci. USA 101, 17711-17715.doi:10.1073/pnas.0407327102

Couso, J. P. (2009). Segmentation, metamerism and the Cambrian explosion.Int. J. Dev. Biol. 53, 1305-1316. doi:10.1387/ijdb.072425jc

Cruz, C., Maegawa, S., Weinberg, E. S., Wilson, S.W., Dawid, I. B. and Kudoh, T.(2010). Induction and patterning of trunk and tail neural ectoderm by thehomeobox gene eve1 in zebrafish embryos. Proc. Natl. Acad. Sci. USA 107,3564-3569. doi:10.1073/pnas.1000389107

Damen, W. G. M. (2002). Parasegmental organization of the spider embryo impliesthat the parasegment is an evolutionary conserved entity in arthropodembryogenesis. Development 129, 1239-1250. doi:10.3410/f.1004866.56706

Damen, W. G. M., Janssen, R. and Prpic, N.-M. (2005). Pair rule gene orthologs inspider segmentation. Evol. Dev. 7, 618-628. doi:10.1111/j.1525-142X.2005.05065.x

Davis, G. K. and Patel, N. H. (2002). Short, long, and beyond: molecular andembryological approaches to insect segmentation. Annu. Rev. Entomol. 47,669-699. doi:10.1146/annurev.ento.47.091201.145251

Davis, G. K., Jaramillo, C. A. and Patel, N. H. (2001). Pax group III genes and theevolution of insect pair-rule patterning. Development 128, 3445-3458.

Davis, G. K., D’Alessio, J. A. and Patel, N. H. (2005). Pax3/7 genes revealconservation and divergence in the arthropod segmentation hierarchy. Dev. Biol.285, 169-184. doi:10.1016/j.ydbio.2005.06.014

Dearden, P. K., Donly, C. and Grbic, M. (2002). Expression of pair-rule genehomologues in a chelicerate: early patterning of the two-spotted spider miteTetranychus urticae. Development 129, 5461-5472. doi:10.1242/dev.00099

Deutsch, J. S. (2004). Segments and parasegments in Arthropods: a functionalperspective. BioEssays 26, 1117-1125. doi:10.1002/bies.20111

DiNardo, S. and O’Farrell, P. H. (1987). Establishment and refinement ofsegmental pattern in the Drosophila embryo: spatial control of engrailedexpression by pair-rule genes. Genes Dev. 1, 1212-1225. doi:10.1101/gad.1.10.1212

Donitz, J., Schmitt-Engel, C., Grossmann, D., Gerischer, L., Tech, M.,Schoppmeier, M., Klingler, M. and Bucher, G. (2015). iBeetle-Base: adatabase for RNAi phenotypes in the red flour beetle Tribolium castaneum.Nucleic Acids Res. 43, D720-D725. doi:10.1093/nar/gku1054

Dray, N., Tessmar-Raible, K., Le Gouar, M., Vibert, L., Christodoulou, F.,Schipany, K., Guillou, A., Zantke, J., Snyman, H., Behague, J. et al. (2010).Hedgehog signaling regulates segment formation in the annelid Platynereis.Science 329, 339-342. doi:10.1126/science.1188913

Eckert, C., Aranda, M., Wolff, C. and Tautz, D. (2004). Separable stripe enhancerelements for the pair-rule gene hairy in the beetle Tribolium. EMBO Rep. 5,638-642. doi:10.1038/sj.embor.7400148

13

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

El-Sherif, E. and Levine, M. (2016). Shadow enhancers mediate dynamic shifts ofgap gene expression in the Drosophila embryo. Curr. Biol. 26, 1164-1169. doi:10.1016/j.cub.2016.02.054

El-Sherif, E., Averof, M. andBrown, S. J. (2012). A segmentation clock operating inblastoderm and germband stages of Tribolium development. Development 139,4341-4346. doi:10.1242/dev.085126

El-Sherif, E., Zhu, X., Fu, J. and Brown, S. J. (2014). Caudal regulates thespatiotemporal dynamics of pair-rule waves in tribolium. PLoS Genet. 10,e1004677. doi:10.1371/journal.pgen.1004677

Elowitz, M. B. and Leibler, S. (2000). A synthetic oscillatory network oftranscriptional regulators. Nature 403, 335. doi:10.1038/35002125

Erezyilmaz, D. F., Kelstrup, H. C. and Riddiford, L. M. (2009). The nuclearreceptor E75A has a novel pair-rule-like function in patterning the milkweed bug,Oncopeltus fasciatus. Dev. Biol. 334, 300-310. doi:10.1016/j.ydbio.2009.06.038

Eriksson, B. J., Tait, N. N., Budd, G. E. and Akam, M. (2009). The involvement ofengrailed and wingless during segmentation in the onychophoran Euperipatoideskanangrensis (Peripatopsidae: Onychophora) (Reid 1996). Dev. Genes Evol.219, 249-264. doi:10.1007/s00427-009-0287-7

Eriksson, B. J., Ungerer, P. and Stollewerk, A. (2013). The function of Notchsignalling in segment formation in the crustacean Daphnia magna(Branchiopoda). Dev. Biol. 383, 321-330. doi:10.1016/j.ydbio.2013.09.021

Farzana, L. and Brown, S. J. (2008). Hedgehog signaling pathway functionconserved in Tribolium segmentation. Dev. Genes Evol. 218, 181-192. doi:10.1007/s00427-008-0207-2

François, P., Hakim, V. and Siggia, E. D. (2007). Deriving structure from evolution:metazoan segmentation. Mol. Syst. Biol. 3, 154. doi:10.1038/msb4100192

French, V. (1988). Gradients and insect segmentation. Development 104, 3-16.Gilles, A. F., Schinko, J. B. and Averof, M. (2015). Efficient CRISPR-mediatedgene targeting and transgene replacement in the beetle Tribolium castaneum.Development 142, 2832-2839. doi:10.1242/dev.125054

Go, M. J., Eastman, D. S. and Artavanis-Tsakonas, S. (1998). Cell proliferationcontrol by Notch signaling in Drosophila development. Development 125,20131-22040.

Goltsev, Y., Hsiong, W., Lanzaro, G. and Levine, M. (2004). Differentcombinations of gap repressors for common stripes in Anopheles andDrosophila embryos. Dev. Biol. 275, 435-446. doi:10.1016/j.ydbio.2004.08.021

Gomez, C., Ozbudak, E. M., Wunderlich, J., Baumann, D., Lewis, J. andPourquie, O. (2008). Control of segment number in vertebrate embryos. Nature454, 335-339. doi:10.1038/nature07020

Graham, A., Butts, T., Lumsden, A. and Kiecker, C. (2014). What can vertebratestell us about segmentation?. EvoDevo 5, 24. doi:10.1186/2041-9139-5-24

Green, J. and Akam, M. (2013). Evolution of the pair rule gene network: Insightsfrom a centipede. Dev. Biol. 382, 235-245. doi:10.1016/j.ydbio.2013.06.017

Griffiths, J. A., Scialdone, A. and Marioni, J. C. (2018). Using single-cellgenomics to understand developmental processes and cell fate decisions. Mol.Syst. Biol. 14, e8046. doi:10.15252/msb.20178046

Hannibal, R. L. and Patel, N. H. (2013). What is a segment? EvoDevo 4, 35. doi:10.1186/2041-9139-4-35

Hannibal, R. L., Price, A. L. and Patel, N. H. (2012). The functional relationshipbetween ectodermal and mesodermal segmentation in the crustacean, Parhyalehawaiensis. Dev. Biol. 361, 427-438. doi:10.1016/j.ydbio.2011.09.033

Hare, E. E., Peterson, B. K., Iyer, V. N., Meier, R. and Eisen, M. B. (2008). Sepsideven-skipped enhancers are functionally conserved in Drosophila despite lack ofsequence conservation. PLoS Genet. 4, e1000106. doi:10.1371/journal.pgen.1000106

He, H. and Noll, M. (2013). Differential and redundant functions of gooseberry andgooseberry neuro in the central nervous system and segmentation of theDrosophila embryo. Dev. Biol. 382, 209-223. doi:10.1016/j.ydbio.2013.05.017

Heffer, A., Grubbs, N., Mahaffey, J. and Pick, L. (2013). The evolving role of theorphan nuclear receptor ftz-f1, a pair-rule segmentation gene. Evol. Dev. 15,406-417. doi:10.1111/ede.12050

Hubaud, A. and Pourquie, O. (2014). Signalling dynamics in vertebratesegmentation. Nat. Rev. Mol. Cell Biol. 15, 709-721. doi:10.1038/nrm3891

Hughes, C. L. and Kaufman, T. C. (2002a). Hox genes and the evolution of thearthropod body plan1. Evol. Dev. 4, 459-499. doi:10.1046/j.1525-142X.2002.02034.x

Hughes, C. L. and Kaufman, T. C. (2002b). Exploring Myriapod Segmentation: TheExpression Patterns of even-skipped, engrailed, and wingless in a Centipede.Dev. Biol. 247, 47-61. doi:10.1006/dbio.2002.0683

Hunding, A. and Baumgartner, S. (2017). Ancient role of ten-m/odz insegmentation and the transition from sequential to syncytial segmentation.Hereditas 154, 8. doi:10.1186/s41065-017-0029-1

Isshiki, T., Pearson, B., Holbrook, S. and Doe, C. Q. (2001). Drosophilaneuroblasts sequentially express transcription factors which specify thetemporal identity of their neuronal progeny. Cell 106, 511-521. doi:10.1016/S0092-8674(01)00465-2

Jacobs, D. K., Hughes, N. C., Fitz-Gibbon, S. T. and Winchell, C. J. (2005).Terminal addition, the Cambrian radiation and the Phanerozoic evolution ofbilaterian form. Evol. Dev. 7, 498-514. doi:10.1111/j.1525-142X.2005.05055.x

Jaeger, J. (2011). The gap gene network. Cell. Mol. Life Sci. 68, 243-274. doi:10.1007/s00018-010-0536-y

Jaeger, J., Surkova, S., Blagov, M., Janssens, H., Kosman, D., Kozlov, K. N.,Manu, Myasnikova, E., Vanario-Alonso, C. E., Samsonova, M. et al. (2004).Dynamic control of positional information in the early Drosophila embryo. Nature430, 368-371. doi:10.1038/nature02678

Janssen, R. (2011). Diplosegmentation in the pill millipede Glomeris marginata isthe result of dorsal fusion.Evol. Dev. 13, 477-487. doi:10.1111/j.1525-142X.2011.00504.x

Janssen, R. and Budd, G. E. (2013). Deciphering the onychophoran ‘segmentationgene cascade’: Gene expression reveals limited involvement of pair rule geneorthologs in segmentation, but a highly conserved segment polarity gene network.Dev. Biol. 382, 224-234. doi:10.1016/j.ydbio.2013.07.010

Janssen, R., Prpic, N.-M. and Damen,W. G. M. (2004). Gene expression suggestsdecoupled dorsal and ventral segmentation in the millipede Glomeris marginata(Myriapoda: Diplopoda). Dev. Biol. 268, 89-104. doi:10.1016/j.ydbio.2003.12.021

Janssen, R., Le Gouar, M., Pechmann, M., Poulin, F., Bolognesi, R., Schwager,E. E., Hopfen, C., Colbourne, J. K., Budd, G. E., Brown, S. J. et al. (2010).Conservation, loss, and redeployment of Wnt ligands in protostomes: implicationsfor understanding the evolution of segment formation. BMC Evol. Biol. 10, 374.doi:10.1186/1471-2148-10-374

Janssen, R., Budd, G. E., Prpic, N.-M. and Damen,W. G. M. (2011). Expression ofmyriapod pair rule gene orthologs. EvoDevo 2, 5. doi:10.1186/2041-9139-2-5

Janssen, R., Damen, W. G. M. and Budd, G. E. (2012). Expression of pair rulegene orthologs in the blastoderm of a myriapod: evidence for pair rule-likemechanisms? BMC Dev. Biol. 12, 15. doi:10.1186/1471-213X-12-15

Janssen, R., Andersson, E., Betner, E., Bijl, S., Fowler, W., Hook, L., Leyhr, J.,Mannelqvist, A., Panara, V., Smith, K. et al. (2018). Embryonic expressionpatterns and phylogenetic analysis of panarthropod sox genes: insight intonervous system development, segmentation and gonadogenesis. BMCEvol. Biol.18, 88. doi:10.1186/s12862-018-1196-z

Jaynes, J. B. and Fujioka, M. (2004). Drawing lines in the sand: even skipped et al.and parasegment boundaries. Dev. Biol. 269, 609-622. doi:10.1016/j.ydbio.2004.03.001

Jin, S., O, J., Stellabotte, F., Brown, S. J. and Choe, C. P. (2019). Expression ofteneurin-m/odd Oz during segmentation in the beetle Tribolium castaneum.GeneExpr. Patterns 31, 26-31. doi:10.1016/j.gep.2019.01.002

Kadner, D. and Stollewerk, A. (2004). Neurogenesis in the chilopod Lithobiusforficatus suggests more similarities to chelicerates than to insects. Dev. GenesEvol. 214, 367-379. doi:10.1007/s00427-004-0419-z

Kainz, F., Ewen-Campen, B., Akam, M. and Extavour, C. G. (2011). Notch/Deltasignalling is not required for segment generation in the basally branching insectGryllus bimaculatus. Development 138, 5015-5026. doi:10.1242/dev.073395

Keranen, S. V., Fowlkes, C. C., Luengo Hendriks, C. L., Sudar, D., Knowles,D. W., Malik, J. and Biggin, M. D. (2006). Three-dimensional morphology andgene expression in the Drosophilablastoderm at cellular resolution II: dynamics.Genome Biol. 7, R124. doi:10.1186/gb-2006-7-12-r124

Kettle, C. and Arthur, W. (2000). Latitudinal cline in segment number in anarthropod species, Strigamia maritima. Proc. R. Soc. Lond. B Biol. Sci. 267,1393-1397. doi:10.1098/rspb.2000.1155

Krause, G. (1939). Die Eitypen der Insekten. Biol. Zentralblatt 59, 495-536.Krol, A. J., Roellig, D., Dequeant, M.-L., Tassy, O., Glynn, E., Hattem, G.,

Mushegian, A., Oates, A. C. and Pourquie, O. (2011). Evolutionary plasticity ofsegmentation clock networks. Development 138, 2783-2792. doi:10.1242/dev.063834

Kux, K., Kiparaki, M. and Delidakis, C. (2013). The two Tribolium E(spl) genesshow evolutionarily conserved expression and function during embryonicneurogenesis. Mech. Dev. 130, 207-225. doi:10.1016/j.mod.2013.02.003

Lai, Y.-T., Deem, K. D., Borras-Castells, F., Sambrani, N., Rudolf, H.,Suryamohan, K., El-Sherif, E., Halfon, M. S., McKay, D. J. and Tomoyasu,Y. (2018). Enhancer identification and activity evaluation in the red flour beetle,Tribolium castaneum. Development 145, dev160663. doi:10.1242/dev.160663

Lauschke, V. M., Tsiairis, C. D., François, P. and Aulehla, A. (2013). Scaling ofembryonic patterning based on phase-gradient encoding. Nature 493, 101-105.doi:10.1038/nature11804

Lewis, J. (2003). Autoinhibition with transcriptional delay: a simple mechanism forthe zebrafish somitogenesis oscillator. Curr. Biol. 13, 1398-1408. doi:10.1016/S0960-9822(03)00534-7

Liao, B.-K. and Oates, A. C. (2017). Delta-Notch signalling in segmentation.Arthropod Struct. Dev. 46, 429-447. doi:10.1016/j.asd.2016.11.007

Lim, B., Fukaya, T., Heist, T. and Levine, M. (2018). Temporal dynamics of pair-rule stripes in living Drosophila embryos. Proc. Natl. Acad. Sci. USA 115,8376-8381. doi:10.1073/pnas.1810430115

Liu, P. Z. and Kaufman, T. C. (2004). hunchback is required for suppression ofabdominal identity, and for proper germband growth and segmentation in theintermediate germband insect Oncopeltus fasciatus. Development 131,1515-1527. doi:10.1242/dev.01046

Liu, P. Z. and Kaufman, T. C. (2005). even-skipped is not a pair-rule gene but hassegmental and gap-like functions in Oncopeltus fasciatus, an intermediategermband insect. Development 132, 2081-2092. doi:10.1242/dev.01807

14

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

Liu, Q., Onal, P., Datta, R. R., Rogers, J. M., Schmidt-Ott, U., Bulyk, M. L., Small,S. and Thornton, J.W. (2018). Ancient mechanisms for the evolution of the bicoidhomeodomain’s function in fly development. eLife 7, e34594. doi:10.7554/eLife.34594

Ludwig, M. Z., Patel, N. H. and Kreitman, M. (1998). Functional analysis of evestripe 2 enhancer evolution in Drosophila: rules governing conservation andchange. Development 125, 949-958.

Lynch, J. A., Brent, A. E., Leaf, D. S., Pultz, M. A. and Desplan, C. (2006).Localized maternal orthodenticle patterns anterior and posterior in the long germwasp Nasonia. Nature 439, 728-732. doi:10.1038/nature04445

Marques-Souza, H., Aranda, M. and Tautz, D. (2008). Delimiting the conservedfeatures of hunchback function for the trunk organization of insects. Development135, 881-888. doi:10.1242/dev.018317

Martin, B. L. and Kimelman, D. (2009). Wnt signaling and the evolution ofembryonic posterior development. Curr. Biol. 19, R215-R219. doi:10.1016/j.cub.2009.01.052

Martin, A., Serano, J. M., Jarvis, E., Bruce, H. S., Wang, J., Ray, S., Barker, C. A.,O’Connell, L. C. and Patel, N. H. (2016). CRISPR/Cas9 mutagenesis revealsversatile roles of Hox genes in crustacean limb specification and evolution. Curr.Biol. 26, 14-26. doi:10.1016/j.cub.2015.11.021

Martinez-Arias, A. and Lawrence, P. A. (1985). Parasegments and compartmentsin the Drosophila embryo. Nature 313, 639-642. doi:10.1038/313639a0

McGregor, A. P. (2005). How to get ahead: the origin, evolution and function ofbicoid. BioEssays 27, 904-913. doi:10.1002/bies.20285

McGregor, A. P., Pechmann, M., Schwager, E. E., Feitosa, N. M., Kruck, S.,Aranda, M. and Damen, W. G. M. (2008). Wnt8 is required for growth-zoneestablishment and development of opisthosomal segments in a spider. Curr. Biol.18, 1619-1623. doi:10.1016/j.cub.2008.08.045

Meinhardt, H. (1982). Models of Biological Pattern Formation. Academic Press.Misof, B., Liu, S., Meusemann, K., Peters, R. S., Donath, A., Mayer, C.,Frandsen, P. B., Ware, J., Flouri, T., Beutel, R. G. et al. (2014). Phylogenomicsresolves the timing and pattern of insect evolution. Science 346, 763-767. doi:10.1126/science.1257570

Mito, T., Sarashina, I., Zhang, H., Iwahashi, A., Okamoto, H., Miyawaki, K.,Shinmyo, Y., Ohuchi, H. and Noji, S. (2005). Non-canonical functions ofhunchback in segment patterning of the intermediate germ cricket Gryllusbimaculatus. Development 132, 2069-2079. doi:10.1242/dev.01784

Mito, T., Kobayashi, C., Sarashina, I., Zhang, H., Shinahara, W., Miyawaki, K.,Shinmyo, Y., Ohuchi, H. and Noji, S. (2007). even-skipped has gap-like, pair-rule-like, and segmental functions in the cricket Gryllus bimaculatus, a basal,intermediate germ insect (Orthoptera). Dev. Biol. 303, 202-213. doi:10.1016/j.ydbio.2006.11.003

Mito, T., Shinmyo, Y., Kurita, K., Nakamura, T., Ohuchi, H. and Noji, S. (2011).Ancestral functions of Delta/Notch signaling in the formation of body and legsegments in the cricket Gryllus bimaculatus. Development 138, 3823-3833.doi:10.1242/dev.060681

Miyawaki, K., Mito, T., Sarashina, I., Zhang, H., Shinmyo, Y., Ohuchi, H. andNoji, S. (2004). Involvement of Wingless/Armadillo signaling in the posteriorsequential segmentation in the cricket, Gryllus bimaculatus (Orthoptera), asrevealed by RNAi analysis. Mech. Dev. 121, 119-130. doi:10.1016/j.mod.2004.01.002

Morelli, L. G., Ares, S., Herrgen, L., Schroter, C., Julicher, F. and Oates, A. C.(2009). Delayed coupling theory of vertebrate segmentation. HFSP J. 3, 55-66.doi:10.2976/1.3027088

Murgan, S., Kari, W., Rothbacher, U., Iche-Torres, M., Melenec, P., Hobert, O.and Bertrand, V. (2015). Atypical transcriptional activation by TCF via a Zictranscription factor in C. elegans neuronal precursors. Dev. Cell 33, 737-745.doi:10.1016/j.devcel.2015.04.018

Nakamoto, A., Hester, S. D., Constantinou, S. J., Blaine, W. G., Tewksbury,A. B., Matei, M. T., Nagy, L. M. and Williams, T. A. (2015). Changing cellbehaviours during beetle embryogenesis correlates with slowing of segmentation.Nat. Commun. 6, 6635. doi:10.1038/ncomms7635

Nakao, H. (2010). Characterization of Bombyx embryo segmentation process:expression profiles of engrailed, even-skipped, caudal, and wnt1/winglesshomologues. J. Exp. Zoolog. B Mol. Dev. Evol. 314B, 224-231. doi:10.1002/jez.b.21328

Nakao, H. (2015). Analyses of interactions among pair-rule genes and the gap geneKruppel in Bombyx segmentation. Dev. Biol. 405, 149-157. doi:10.1016/j.ydbio.2015.06.012

Nakao, H. (2016). Hunchback knockdown induces supernumerary segmentformation in Bombyx. Dev. Biol. 413, 207-216. doi:10.1016/j.ydbio.2016.03.024

Nakao, H. (2018). A Bombyx homolog of ovo is a segmentation gene that actsdownstream of Bm-wnt1 (Bombyx wnt1 homolog). Gene Expr. Patterns 27, 1-7.doi:10.1016/j.gep.2017.10.002

Nasiadka, A., Dietrich, B. H. and Krause, H. M. (2002). Anterior-posteriorpatterning in the Drosophila embryo. In Advances in Developmental Biology andBiochemistry, Gene Expression at the Beginning of Animal Development (ed. M.L. DePamphilis), pp. 155-204. Elsevier.

Nusslein-Volhard, C. and Wieschaus, E. (1980). Mutations affecting segmentnumber and polarity in Drosophila. Nature 287, 795. doi:10.1038/287795a0

Oates, A. C., Morelli, L. G. and Ares, S. (2012). Patterning embryos withoscillations: structure, function and dynamics of the vertebrate segmentationclock. Development 139, 625-639. doi:10.1242/dev.063735

Oberhofer, G., Grossmann, D., Siemanowski, J. L., Beissbarth, T. and Bucher,G. (2014). Wnt/β-catenin signaling integrates patterning and metabolism of theinsect growth zone. Development 141, 4740-4750. doi:10.1242/dev.112797

Oda, H., Nishimura, O., Hirao, Y., Tarui, H., Agata, K. and Akiyama-Oda, Y.(2007). Progressive activation of Delta-Notch signaling from around theblastopore is required to set up a functional caudal lobe in the spiderAchaearanea tepidariorum. Development 134, 2195-2205. doi:10.1242/dev.004598

Paese, C. L. B., Schoenauer, A., Leite, D. J., Russell, S. and McGregor, A. P.(2018). A SoxB gene acts as an anterior gap gene and regulates posteriorsegment addition in a spider. eLife 7, e37567. doi:10.7554/eLife.37567

Pare, A. C., Vichas, A., Fincher, C. T., Mirman, Z., Farrell, D. L., Mainieri, A. andZallen, J. A. (2014). A positional Toll receptor code directs convergent extensionin Drosophila. Nature 515, 523-527. doi:10.1038/nature13953

Patel, N. H., Condron, B. G. and Zinn, K. (1994). Pair-rule expression patterns ofeven-skipped are found in both short- and long-germ beetles. Nature 367, 429.doi:10.1038/367429a0

Pechmann, M., Khadjeh, S., Turetzek, N., McGregor, A. P., Damen, W. G. M. andPrpic, N.-M. (2011). Novel function of distal-less as a gap gene during spidersegmentation. PLoS Genet. 7, e1002342. doi:10.1371/journal.pgen.1002342

Peel, A. and Akam, M. (2003). Evolution of segmentation: rolling back the clock.Curr. Biol. 13, R708-R710. doi:10.1016/j.cub.2003.08.045

Peel, A. D., Chipman, A. D. andAkam,M. (2005). Arthropod segmentation: beyondthe Drosophila paradigm. Nat. Rev. Genet. 6, 905-916. doi:10.1038/nrg1724

Pick, L. (2016). Hox genes, evo-devo, and the case of the ftz gene. Chromosoma125, 535-551. doi:10.1007/s00412-015-0553-6

Posnien, N., Schinko, J. B., Kittelmann, S. and Bucher, G. (2010). Genetics,development and composition of the insect head – a beetle’s view. ArthropodStruct. Dev. Evol. Pattern. Mech. 39, 399-410. doi:10.1016/j.asd.2010.08.002

Pourebrahim, R., Houtmeyers, R., Ghogomu, S., Janssens, S., Thelie, A., Tran,H. T., Langenberg, T., Vleminckx, K., Bellefroid, E., Cassiman, J.-J. et al.(2011). Transcription factor Zic2 inhibits Wnt/β-catenin protein signaling. J. Biol.Chem. 286, 37732-37740. doi:10.1074/jbc.M111.242826

Prud’homme, B., de Rosa, R., Arendt, D., Julien, J.-F., Pajaziti, R., Dorresteijn,A. W. C., Adoutte, A., Wittbrodt, J. and Balavoine, G. (2003). Arthropod-likeexpression patterns of engrailed and wingless in the annelid Platynereis dumeriliisuggest a role in segment formation.Curr. Biol. 13, 1876-1881. doi:10.1016/j.cub.2003.10.006

Pueyo, J. I., Lanfear, R. and Couso, J. P. (2008). Ancestral Notch-mediatedsegmentation revealed in the cockroach Periplaneta americana. Proc. Natl. Acad.Sci. USA 105, 16614-16619. doi:10.1073/pnas.0804093105

Reding, K., Chen, M., Lu, Y., Cheatle Jarvela, A. M. and Pick, L. (2019).Shifting roles of Drosophila pair-rule gene orthologs: segmental expression andfunction in the milkweed bug Oncopeltus fasciatus. Development 146,dev181453. doi:10.1242/dev.181453

Richmond, D. L. and Oates, A. C. (2012). The segmentation clock: inherited trait oruniversal design principle? Curr. Opin. Genet. Dev. 22, 600-606. doi:10.1016/j.gde.2012.10.003

Rohr, K. B., Tautz, D. and Sander, K. (1999). Segmentation gene expression in themothmidge Clogmia albipunctata (Diptera, psychodidae) and other primitivedipterans. Dev. Genes Evol. 209, 145-154. doi:10.1007/s004270050238

Rosenberg, M. I., Brent, A. E., Payre, F. and Desplan, C. (2014). Dual mode ofembryonic development is highlighted by expression and function of Nasonia pair-rule genes. eLife 3, e01440. doi:10.7554/eLife.01440

Rothschild, J. B., Tsimiklis, P., Siggia, E. D. and François, P. (2016). Predictingancestral segmentation phenotypes from Drosophila to anopheles using in silicoevolution. PLoS Genet. 12, e1006052. doi:10.1371/journal.pgen.1006052

Salazar-Ciudad, I., Newman, S. A. and Sole, R. V. (2001). Phenotypic anddynamical transitions in model genetic networks I. Emergence of patterns andgenotype-phenotype relationships. Evol. Dev. 3, 84-94. doi:10.1046/j.1525-142x.2001.003002084.x

Samee, M. A. H., Lydiard-Martin, T., Biette, K. M., Vincent, B. J., Bragdon, M. D.,Eckenrode, K. B., Wunderlich, Z., Estrada, J., Sinha, S. and DePace, A. H.(2017). Quantitative measurement and thermodynamic modeling of fusedenhancers support a two-tiered mechanism for interpreting regulatory DNA. CellRep. 21, 236-245. doi:10.1016/j.celrep.2017.09.033

Sander, K. (1976). Specification of the basic body pattern in insect embryogenesis.Adv. Insect Physiol. 12, 125-238. doi:10.1016/S0065-2806(08)60255-6

Sarrazin, A. F., Peel, A. D. and Averof, M. (2012). A segmentation clock with two-segment periodicity in insects. Science 336, 338-341. doi:10.1126/science.1218256

Scholtz, G. (1992). Cell lineage studies in the crayfish Cherax destructor(Crustacea, Decapoda): germ band formation, segmentation, and earlyneurogenesis. Roux Arch. Dev Biol 202, 36-48. doi:10.1007/BF00364595

Schonauer, A., Paese, C. L. B., Hilbrant, M., Leite, D. J., Schwager, E. E.,Feitosa, N. M., Eibner, C., Damen, W. G. M. and McGregor, A. P. (2016). TheWnt and Delta-Notch signalling pathways interact to direct pair-rule gene

15

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT

expression via caudal during segment addition in the spider Parasteatodatepidariorum. Development 143, 2455-2463. doi:10.1242/dev.131656

Schoppmeier, M. andDamen,W. G. M. (2005a). Expression of Pax group III genessuggests a single-segmental periodicity for opisthosomal segment patterning inthe spider Cupiennius salei. Evol. Dev. 7, 160-169. doi:10.1111/j.1525-142X.2005.05018.x

Schoppmeier, M. and Damen, W. G. M. (2005b). Suppressor of Hairless andPresenilin phenotypes imply involvement of canonical Notch-signalling insegmentation of the spider Cupiennius salei. Dev. Biol. 280, 211-224. doi:10.1016/j.ydbio.2005.01.024

Schroeder, M. D., Greer, C. and Gaul, U. (2011). How to make stripes: decipheringthe transition from non-periodic to periodic patterns in Drosophila segmentation.Development 138, 3067-3078. doi:10.1242/dev.062141

Schroter, C., Ares, S., Morelli, L. G., Isakova, A., Hens, K., Soroldoni, D.,Gajewski, M., Julicher, F., Maerkl, S. J., Deplancke, B. et al. (2012). Topologyand dynamics of the Zebrafish segmentation clock core circuit. PLoS Biol. 10,e1001364. doi:10.1371/journal.pbio.1001364

Schulz, C., Schroder, R., Hausdorf, B., Wolff, C. and Tautz, D. (1998). A caudalhomologue in the short germ band beetle Tribolium shows similarities to both, theDrosophila and the vertebrate caudal expression patterns. Dev. Genes Evol. 208,283-289. doi:10.1007/s004270050183

Schwager, E. (2008). Segmentation of the spider Achaearanea tepidarioruminvestigated by gene expression and functional analysis of the gap genehunchback. PhD thesis, Universitat zu Koln.

Schwager, E. E., Pechmann, M., Feitosa, N. M., McGregor, A. P. and Damen,W. G. M. (2009). hunchback functions as a segmentation gene in the spiderAchaearanea tepidariorum. Curr. Biol. CB 19, 1333-1340. doi:10.1016/j.cub.2009.06.061

Schwentner, M., Combosch, D. J., Nelson, J. P. and Giribet, G. (2017). Aphylogenomic solution to the origin of insects by resolving crustacean-hexapodrelationships. Curr. Biol. 27, 1818-1824.e5. doi:10.1016/j.cub.2017.05.040

Seaver, E. C. and Kaneshige, L. M. (2006). Expression of “segmentation” genesduring larval and juvenile development in the polychaetes Capitella sp. I andH. elegans. Dev. Biol. 289, 179-194. doi:10.1016/j.ydbio.2005.10.025

Seaver, E. C., Paulson, D. A., Irvine, S. Q. and Martindale, M. Q. (2001). Thespatial and temporal expression of Ch-en, the engrailed gene in the polychaeteChaetopterus, does not support a role in body axis segmentation. Dev. Biol. 236,195-209. doi:10.1006/dbio.2001.0309

Sharpe, J. (2017). Computer modeling in developmental biology: growing today,essential tomorrow. Development 144, 4214-4225. doi:10.1242/dev.151274

Shinmyo, Y., Mito, T., Matsushita, T., Sarashina, I., Miyawaki, K., Ohuchi, H. andNoji, S. (2005). caudal is required for gnathal and thoracic patterning and forposterior elongation in the intermediate-germband cricket Gryllus bimaculatus.Mech. Dev. 122, 231-239. doi:10.1016/j.mod.2004.10.001

Soroldoni, D., Jorg, D. J., Morelli, L. G., Richmond, D. L., Schindelin, J.,Julicher, F. and Oates, A. C. (2014). A Doppler effect in embryonic patternformation. Science 345, 222-225. doi:10.1126/science.1253089

Stahi, R. and Chipman, A. D. (2016). Blastoderm segmentation in Oncopeltusfasciatus and the evolution of insect segmentation mechanisms. Proc. R. Soc. BBiol. Sci. 283, 20161745. doi:10.1098/rspb.2016.1745

Steventon, B., Duarte, F., Lagadec, R., Mazan, S., Nicolas, J.-F. and Hirsinger,E. (2016). Species-specific contribution of volumetric growth and tissueconvergence to posterior body elongation in vertebrates. Development 143,1732-1741. doi:10.1242/dev.126375

Stollewerk, A., Schoppmeier, M. and Damen, W. G. M. (2003). Involvement ofNotch and Delta genes in spider segmentation. Nature 423, 863-865. doi:10.1038/nature01682

Stuart, J. J., Brown, S. J., Beeman, R. W. and Denell, R. E. (1991). A deficiency ofthe homeotic complex of the beetle Tribolium. Nature 350, 72. doi:10.1038/350072a0

Sucena, É., Vanderberghe, K., Zhurov, V. and Grbic, M. (2014). Reversion ofdevelopmental mode in insects: evolution from long germband to short germbandin the polyembrionic wasp Macrocentrus cingulum Brischke. Evol. Dev. 16,233-246. doi:10.1111/ede.12086

Surkova, S., Kosman, D., Kozlov, K., Manu, Myasnikova, E., Samsonova, A. A.,Spirov, A., Vanario-Alonso, C. E., Samsonova, M. and Reinitz, J. (2008).Characterization of the Drosophila segment determination morphome. Dev. Biol.313, 844-862. doi:10.1016/j.ydbio.2007.10.037

Vedel, V., Chipman, A. D., Akam, M. and Arthur, W. (2008). Temperature-dependent plasticity of segment number in an arthropod species: the centipedeStrigamia maritima. Evol. Dev. 10, 487-492. doi:10.1111/j.1525-142X.2008.00259.x

Vedel, V., Apostolou, Z., Arthur, W., Akam, M. and Brena, C. (2010). An earlytemperature-sensitive period for the plasticity of segment number in the centipedeStrigamia maritima. Evol. Dev. 12, 347-352. doi:10.1111/j.1525-142X.2010.00421.x

Vellutini, B. C. and Hejnol, A. (2016). Expression of segment polarity genes inbrachiopods supports a non-segmental ancestral role of engrailed for bilaterians.Sci. Rep. 6, 32387. doi:10.1038/srep32387

Verd, B., Crombach, A. and Jaeger, J. (2014). Classification of transientbehaviours in a time-dependent toggle switch model. BMC Syst. Biol. 8, 43.doi:10.1186/1752-0509-8-43

Verd, B., Clark, E., Wotton, K. R., Janssens, H., Jimenez-Guri, E., Crombach, A.and Jaeger, J. (2018). A damped oscillator imposes temporal order on posteriorgap gene expression in Drosophila. PLoS Biol. 16, e2003174. doi:10.1371/journal.pbio.2003174

Vroomans, R. M. A., Hogeweg, P. and Ten Tusscher, K. H. W. J. (2016). In silicoevo-devo: reconstructing stages in the evolution of animal segmentation.EvoDevo 7, 14. doi:10.1186/s13227-016-0052-8

Vroomans, R. M. A., Hogeweg, P. and ten Tusscher, K. H. W. J. (2018). Aroundthe clock: gradient shape and noise impact the evolution of oscillatorysegmentation dynamics. EvoDevo 9, 24. doi:10.1186/s13227-018-0113-2

Williams, T. A. and Nagy, L. M. (2017). Linking gene regulation to cell behaviors inthe posterior growth zone of sequentially segmenting arthropods. ArthropodStruct. Dev. 46, 380-394. doi:10.1016/j.asd.2016.10.003

Williams, T., Blachuta, B., Hegna, T. A. and Nagy, L. M. (2012). Decouplingelongation and segmentation: notch involvement in anostracan crustaceansegmentation. Evol. Dev. 14, 372-382. doi:10.1111/j.1525-142X.2012.00555.x

Wilson, M. J., McKelvey, B. H., van der Heide, S. and Dearden, P. K. (2010).Notch signaling does not regulate segmentation in the honeybee, Apis mellifera.Dev. Genes Evol. 220, 179-190. doi:10.1007/s00427-010-0340-6

Wolff, C., Tinevez, J.-Y., Pietzsch, T., Stamataki, E., Harich, B., Guignard, L.,Preibisch, S., Shorte, S., Keller, P. J., Tomancak, P. et al. (2018). Multi-viewlight-sheet imaging and tracking with the MaMuT software reveals the cell lineageof a direct developing arthropod limb. eLife 7, e34410. doi:10.7554/eLife.34410

Xiang, J., Reding, K., Heffer, A. and Pick, L. (2017). Conservation and variation inpair-rule gene expression and function in the intermediate-germ beetle Dermestesmaculatus. Development 144, 4625-4636. doi:10.1242/dev.154039

Xu, J. and Gridley, T. (2012). Notch signalling during oogenesis in Drosophilamelanogaster. Genet. Res. Int. 2012, 648207. doi:10.1155/2012/648207

Zheng, L., Michelson, Y., Freger, V., Avraham, Z., Venken, K. J. T., Bellen, H. J.,Justice, M. J. and Wides, R. (2011). Drosophila Ten-m and filamin affect motorneuron growth cone guidance. PLoS ONE 6, e22956. doi:10.1371/journal.pone.0022956

Zhu, X., Rudolf, H., Healey, L., François, P., Brown, S. J., Klingler, M. and El-Sherif, E. (2017). Speed regulation of genetic cascades allows for evolvability inthe body plan specification of insects. Proc. Natl. Acad. Sci. USA 2017,E8646-E8655. doi:10.1073/pnas.1702478114

16

REVIEW Development (2019) 146, dev170480. doi:10.1242/dev.170480

DEVELO

PM

ENT